I’m really surprised by the resiliency shown by the two TransCanada PipeLines issues – these are very similar perpetuals, with a $50 par value and pay $2.80 p.a. – a coupon of 5.6%. TCA.PR.X is redeemable at par commencing 2013-10-15, while the TCA.PR.Y is redeemable at par commencing 2014-3-5.
TCA.PR.X was issued in October 1998 as TRP.PR.X, while TCA.PR.Y began life 1999-3-5 as TRP.PR.Y. Four million shares of each series are outstanding so they’re a nice size for non-financial issues.
These issues are perennial favourites of mine. They were hard hit when TRP cut its common share dividend, with the low point being 2000-5-23: TRP.PR.Y had closing quote of 34.80-25 on volume of 6,180 shares. I made a fair bit of money on that – tough times do not lead inevitably to default.
There were some credit worries when they made a big investment in Dec 06, but these were taken care of by an equity issue.
More recently, their 5.6% coupon, far higher than most of their competition in recent years (other issues with similarly high coupons have been called) made them exemplars of the virtues of the PerpetualPremium class – when they yielded 4.10% to call, as they did about a year ago, the difference between this yield and the coupon implied a lot of interest-rate protection for investors.
They’ve weathered the storm of the past year beautifully – well down from the high of 55.71-10 on no volume, reached 2006-12-4 by TCA.PR.Y, but not nearly as badly hit as perpetuals without such high coupons … just chugging along, paying their coupon, and still trading above thier call price.
Which is my problem. Why are they still trading above their call price? The cycle has turned, and a coupon of 5.6% is not as extraordinary as it was a year ago – see the new issues of TD.PR.R, TD.PR.Q and BNS.PR.O all with similar coupons and a call date at par that is further away than the TCA calls (and it is unequivocally better for the call date to be further away, since the call won’t be exercised if you want it to be – and vice versa!).
Why are TCA.PR.X and TCA.PR.Y, both rated Pfd-2(low) by DBRS and P-2 by S&P, trading to yield less than the bank issues, rated Pfd-1 [DBRS] and P-1(low) [S&P]? One explanation may be scarcity value (many players are fully loaded on banks in general and these banks in particular) and another might be extreme sector aversion to financials. But it still doesn’t make a lot of sense to me.
I’ve uploaded some charts, comparing these two issues with others that have a 5.6% coupon…
- YTW TCA.PR.X & W.PR.J
- Modified Duration TCA.PR.X & W.PR.J
- Yield Difference TCA.PR.X & W.PR.J
- Yield Disparity TCA.PR.X
- YTW, CM.PR.E & POW.PR.A
- Modified Duration, CM.PR.E & POW.PR.A
- YTW Difference, CM.PR.E & POW.PR.A
Yield disparity, by the way, is the amount of yield that would have to be added or subtracted from the yield curve in order to achieve a calculated price equal to the market price – some players may know this as the “Z-Spread”. It is not unusual for an issue (such as TCA.PR.X over the past year) to be “always expensive” – this may mean that there is something about the issue that is not incorporated in the model (a restrictive covenant, perhaps, or scarcity value, or … something) but it is clear to see from the chart that TCA.PR.X (and TCA.PR.Y) have become more expensive than usual.
And I completely fail to understand why they’re trading through the banks.
Update, 2008-03-23: In response to prefhound‘s points in the comments, I have uploaded listings for PerpetualDiscount and PerpetualPremium yieldDisparities. Note that these yield disparities contain adjustments for Cumulative Dividends – which I believe to be an artefact, but will admit that I am unsure. The cumulativeDividend adjustment to curve price (and hence curve yield) is quite substantial – without it, TCA.PR.X would appear even more expensive than they do now.
Three reasons:
1. They are not financials. Compare bank bonds and prefs and TRP bonds and prefs — not much change in the yields of boring old utilities. Utilities don’t have leverage as high as banks and financials. Hard assets may be preferable to financial assets these days.
2. Utilities have simple financial statements with pretty clear assets and liabilities. Not as many questions about transparency, mark-to-market or where something could blow up unexpectedly.
3. There aren’t many utility prefs (a few FTS, CU, the odd ENB), and investors who need diversification should love this sector. Infrastructure has a pretty good looking future, in spite of the credit crunch. I suspect the whole utility sector has done way better than banks, not just TRP.
Also, when (and it will be when, not if) spreads return to normal, utility prefs will probably not bounce back as strongly as bank prefs.
I’ve done a bit of bond research this weekend to check this and other spreads:
May 11, 2007 TD Bank Duration 5 bond spread to Canadas was 40 bp
TRP Duration 7.4 bond spread was 80 bp
Mar 20, 2008 TD Bank Duration 4 bond spread was 165 bp (up 125 bp) and
Duration 8 spread was 240 bp
TRP Duration 4 bond spread was 100 bp Duration 8 was 165 bp
(a rise of perhaps 80 bp)
Conclusion: Despite a lower rating, TRP bond AND pref spreads have widened less than spreads for banks and financials. Enbridge is similar. This difference may not reverse until the credit crunch dissipates.
What issue of TD Bank did you use? Sub-Debt is not strictly comparable to straight debt; Sub-debt spreads have widened considerably more than deposit notes; less than Innovative Tier 1 Capital.
I am aware that:
I have attached to the main post an update showing the two perpetuals sub-indices and their yield disparites – it doesn’t seem entirely cut-and-dried!
I’d be the first to admit that my bond info sources are weak (Globe and Canadianfixedincome.ca).
The TD Bond of May 11/07 was described as 5.69 coupon maturing June 3, 2018 for a yield of 4.62% at a price of 105.56. The TRP bond (I own) was 6.50 maturing Dec 9, 2030 yielding 7.25% when at 91.113. I’m not sure whether the bank debt so described is sub or straight, but I suspect straight.
For Mar 20/08 prices, I made graphs of spreads vs duration for TRP and TD with all available bonds on Canadianfixedincome.ca, performed linear regressions and read off spreads from there (only if they were interpolations or short extrapolations). Again, I am assuming all bank bonds so listed are straight. If they are a mixture of straight and sub, there wasn’t much difference.
In your curve prices, your model has no industry parameter, which over time should average out to zero. You use only credit parameters. Right now (and in any stress situations when spreads widen) we are in a time when supply and demand result in wider spreads for better credits, particularly if they are financials.
I don’t think CDS are a quantitatively reasonable predictor of the actual future losses. When I have analyzed S&P data on spreads, default rates, recoveries and credit ratings, I figure that when spreads are at their lowest, they still overestimate the historically-based expected cost of default by at least a factor of 2X. When spreads are as high as they are now, this overestimate is 10X or more. I can rationalize 2X as fair compensation for risk under an equilibrium model. 10X just shows how supply and demand can drive markets far from a rational equilibrium in the short-medium term.
Credit ratings are supposed to look through the business cycle, which would say that — if the financial credit ratings are correct — then higher rated banks should see lower spreads than TRP when averaged over the business cycle. Let’s look at TRP vs banks in 12 and 24 months.
The TD bond you looked at is sub-debt – these have un-bond-like characteristics, not so un-bond-like as perpetual prefs, but are not quite equivalent to regular debt.
You are correct, the model I use to fit the curve does not include a parameter to indicate industry; nor is there one to indicate the leverage of the credit. I have a certain amount of the programming done to add a parameter that will account for trends in the underlying equity, but the programming hasn’t been finished yet, let alone tested!
The spread on investment grade bonds is largely due to considerations of liquidity; the actual risk of default is only part of the answer … at least in normal times!
Hey guys . . . I think you both got it bang on with the explanation that these prefs are simply not financials. Unless I’m mistaken, there’s not a financial pref IPO in the last five years that didn’t debut, and in 90% of the cases, stay, underwater. The banks, and the insurance companies have crafted a reputation of printing pref stock like monopoly money; amazingly the market keeps eating it up, and absorbing the losses. Unfortunately, the secondary market is less accomodative.
Transcanada, Westcost, Union Gas are all examples of companies that treat preferred share issues like organized investment vehicles that have actually had some thought put behind them. The market applauds this kind of prudence, and rewards these issues with an appropriate support level for their stock.
As I see the nonsense of Bear Stearns/JP Morgan unfolding, $2/share . . . no we really meant $10/share . . . just kidding! . . . I realize that the financial issuers probably deserve the sorry state they find themselves in now. Unfortunately for the billion dollar per quarter-earning Canadian banks, even a histeria-based rally such as the one continuing today can’t do anything for the lack of confidence establised in financial prefs. 25,000 shares of SLF.PR.C just crossed at $20.50, down .20 to yield 5.46% with a possible redemption premium of $4.50/share.
When it’s over, it’s over.
madequota
Well, fair enough – if you want to say that the entire difference is explained by the fact that TCA is a non-financial company, that’s fine. You’re entitled to your opinions.
But I’m showing W.PR.H & W.PR.J (yields 5.85% and 5.92%, respectively) as being fairly priced, and CIU.PR.A (yield 5.64%) & FAL.PR.H (analyzed as an imminent call) as being cheap … only ENB.PR.A (yield 5.58%) is comparably (although not so much) expensive.
You can say HIMIPref™ is a load of hokum and the rich/cheap analysis is random if you like … but I don’t get it.
HIMIPref is not a load of hokum. Far from it. As a matter of fact, having been involved in this particular category probably way longer than any sane investor should be, I was actually delighted to come across the whole HIMI range of websites and products (and this blog!). When it comes to academic-level detail and analysis of preferred investing, I suspect there is no more valuable a resource available anywhere. Beyond that, the level of commentary here, and the high quality of responses by all participants is far more sophisticated than we generally see on the internet . . . or anywhere for that matter.
Now, having said that, my comments about the current pref situation created primarily by the big 5 banks, but also other financials is a problem . . . a big problem. They really have continued to flood the market with reckless abandon, and it’s showing up in the relative demand/yield comparisons. Your example of TCA is a perfect example of what I suspect is rusting away the preferred market (for the financial issuers anyway). It’s simple supply/demand economics. The banks over the past 6 months have ensured that supply easily outstrips demand, and whenever we see a rally developing . . . well here come the 5.3’s . . . here come the 5.6’s . . . if TD.PR.R gets a little closer to Q above par . . . well here come the 5.7’s.
Sometimes, all the analysis in the world, even high quality analysis, can be rendered neutral by the, unfortunately, predictable bad behaviour of the issuers . . . but it will all end, and the analysis will once again apply to the entire market! I’m just having a hard time being patient with these guys.
madequota
Thanks for the kind words, madequota!
Anyway, I have singled out Brookfield issues (BAM) as having saturated the market for a single name, particularly since their horrendous performance in August 2007. Would it be fair to summarize your remarks as a view that a non-financial/financial discrepency is not so much due to perceptions of credit risk as it is to market saturation?
That is, to the extent that there actually is a financial/non-financial discrepency! While I agree that this is part of the answer, I remain unconvinced that it exists to a large and measurable systemic extent.
absolutely; I don’t think there’s many investors out there who believe any of the Canadian banks’ pref dividends are in jeopardy [regardless of the varying unknowns related to ABCP] . . . but there are many investors [and by now, you know I’m one of them] that are more concerned about the saturation issue continuing.
I’d love to go long some TD.PR.R under par, and enjoy the 5.6%+ dividend in what I believe is a declining interest rate environment. But I’m hesitant, because when RBC or somebody else comes to market next week with a whack of 5.7’s or 5.8’s, TD.PR.R loses a half buck a share, and I’m stuck with it (kinda like TD.PR.Q the day TD.PR.R was announced).
The only sort of safe play right now, I think, are any number of the fiprefs trading in the $20 – $22 range. Even though they also tank as saturation continues, somewhere down the road, you might be rewarded with a big take-out premium. Also, the past 6 months of history seems to show $20 as a support level for all of these financial-issued preferreds . . . and we’re pretty much back to those levels right now with some of them.
I suspect history will prove that this was an amazing time to buy and hold this stuff . . . unfortunately right now, one must buy and hold very selectively!
madequota
I would also add that WN prefs are “distressed” in their own way — falling profitability at Loblaws. Loblaws and Weston debt spreads are also wide, so again, pref and bond spreads are qualitatively consistent.
This leaves only ENB and CIU prefs as direct comparables. You have a Nov 6 post showing ENB.PR.A was occasionally trading at negative YTW — which would imply an “expensive” valuation (though maybe not many cents). CIU pref is not that actively traded.
For now, industry seems the best explanation for TCA pref prices. We also have to ask if the liquidity variable in your model is independent of industry and/or credit rating. Right now TCA is “not well explained by the assumed model”
prefhound … I referred to “W” for “Westcoast”, not “WN” for “Weston”.
I suggest that TCA is explained perfectly by the assumed model … and it’s rich! Whether or not the assumed model is accurate in this evaluation is, of course, question #2!
Oooops! My error on W vs WN. W is not too liquid either.
OK, so you attribute all unexplained variation to “richness” and “cheapness”. If you are right, then TCA prefs should underperform the market (and especially financials) looking forward.
However, as an investor long TCA prefs, I want to maintain the diversification value and might even buy more and be happy with my 5.5% yield (7+% interest equivalent).
To take advantage of your richness, I would need to sell TCA prefs and buy financial prefs, thereby giving up my diversification value (which has already worked demonstrably well in this case). Because I (and perhaps other pref investors) am reluctant to do this, TCA prefs may remain rich until the market recovers.
On the other hand, if your model is correct, TCA prefs may drift downward until they are less rich, even if the pref market as a whole does nothing.
I’m willing to bet a fall lunch that the first scenario is more likely.
prefhound
OK, then it looks to me like I can summarize your view similarly to madequota‘s … that you, personally, are saturated with financials and therefore you’re going to retain your non-financial.
Am I accurate so far?
You could retain non-financial exposure and – according to me – pick up value by selling TCA and buying either CIU or FAL. Even W, for all its relative illiquidity, is better value than TCA – according to me. Is there any reason why you would not do this?
Oops, how could I fail to mention … CU.PR.A (yielding 5.89%) and CU.PR.B (yielding 5.86%) are also cheap … according to HIMIPref™ anyway!
I think ph is moving in the right direction with this. I would go as far as arguing that the fiprefs should probably be dropped at least a notch into the P2 category now; maybe some of them even to P3, but that’s another explanation.
I wanted to share a few thoughts about CIU.PR.A. As you’re aware, this thing came to market last April I believe, shortly after CU redeemed 3 separate pref issues that were quite old, and quite high yielding. I suspect that many of these holders simply re-invested their redemption cash in CIU.PR.A, and now that it’s acting like a sunk fipref, they have no choice but to stay in it, collect divs, and wait.
Interesting point is that, other than the market maker, about the only bidder I’ve ever seen on this one is BMO, using the infamous iceberg buying method! They change their price point generally every week, to reflect any trading activity in the previous week.
Right now, they’re at 20.55, but there’s been some trading activity between 20.70 and 21.00, and the market maker’s come up with 3 separate bids above BMO’s iceberg. If history is correct, we’re about to see a new BMO [buy order] iceberg, probably at $20.90 . . . maybe even $21 since there’s absolutely nothing being offerred.
So ph, if you want to take jh’s advice, and buy some of this one . . . it could be challenging!
madequota
Actually, the redemption of the three older CU issues was shortly after the CIU.PR.A issue, but the redemptions were inevitable.
I agree that volume is quite low. One cause might be their inclusion in the index last July; another might be that nobody wants to sell a high-quality utility issue.
I will note that I haven’t actually gone so far as to advise people to rush out and buy the issue … what I have done is pointed out that it appears cheap to me relative to TCA.PR.X and in the absence of dynamic factors, and all this in the context of providing a counter-example to the non-financials sectoral theory of why the TCA issues are priced where they are.
OK, so CIU and W are out on liquidity basis, but CU.PR.A and CU.PR.B are only 20-30 bp more than TCA prefs. I don’t usually switch for 30 bp, what with bid-ask, differeing ex div dates and other costs (see calc below). I like to switch for maybe 50+ bp (or in the case of BNA.PR.B/C 100+ bp) when the aggravation is worth it. Just like I want at least $1.00 in an arbitrage trade (which might take 3 weeks or 7 months!), there is a law of diminishing returns. I want my trading to return $300/hour for the effort, on top of the income.
Example: Suppose I want to sell $30K face of TCA.PR.Y in favour of a CU pref. Here’s the math based on Level II quotes at 2:40 pm today:
Balancing last trade and bid-ask, I get the following “most reasonable” prices:
TCA.PR.Y: $50.68 to yield 5.60% (to a $50 call in 2014)
CIU.PR.A: $20.62 to yield 5.60%
CU.PR.A: $25.09 to yield 5.82% (to a $25 call in 2011)
CU.PR.B: $25.20 to yield 6.00% (to a $25 call in 2012)
Looks (on paper) like I could gain 40 bp switching from TCA.PR.Y to CU.PR.B, which would be $120/year for 5-6 years (BEFORE COSTS).
Now, let’s look at what I can actually do with these trades (assuming $10 comm)
I can readily sell 600 TCA.PR.Y at the bid (700 @ $50.50) for $50.485 net to give up a yield of 5.62%.
To invest $30K in CU.PR.B, I find only 900 shares on offer at prices from $25.30 to 25.53 (of which only 200 are below $25.43). If I’m lucky, I might buy the needed 1200 shares at an average price of $25.51 net of commission, for a realized yield of 5.68%, not the $6.0% I had in mind. Ooops, 40 bp apparent, just turned into 6 bp actual!
The situation with CU.PR.A is slightly better as I only need to take out 4 sell orders to get 1200 shares, for a net price of $25.16 and yield of 5.81% — a whole 19 bp better, or $57/year. It is not worth it.
I would also add that TCA.Pr.Y liquidity these days is better (average volume of $80,000 per day over the past 24 trading days) than CU.Pr.B or .A ($48,000 or $35,000, respectively), so TCA.PR.Y deserves a slightly lower yield.
Based on this, I’m going to keep my 50 bp minimum criterion for switching, and continue to argue TCA prefs are in little danger of correcting to $48. If I have some EXTRA funds for prefs, I can try to buy CU.PR.A for 5.8%, but may prefer the liquidity of the new 5.8% BMO issue (I have no BMO prefs).
One reason that your multi-parameter model has generated the (apparently false for the real world) TCA switching signal is that you are basing prices on the bids.
Finally, congratulations to madequota on his prescience in forecasting the BMO new issue, which has already garnered its own 17 comments! I would have bought BMO pref in the offering, perhaps even with borrowed money, but it sold out while I was out hitting golf balls 😉
Oh, I wasn’t out-and-out recommending a switch – frictional effects are explicitly accounted for in HIMIPref™ and no such signal has been generated. I really intended to point out that – according to me – there appears to better value in the non-financials sector than is available via an investment in TCA. So why are people bidding for TCA rather than bidding for CU / CIU?
I bring up the question of relative value as a counterpoint to the argument that TCA issues are benefitting from a general sectoral shift into utilities away from financials; the CU / CIU data does not appear to bear this out, particularly if you are influenced by the relative DBRS credit rankings (CU / CIU is ranked two notches better than TCA) or by S&P’s credit evaluation (CU / CIU is one notch better) … and HIMIPref™ is, in fact, heavily influenced by credit considerations, especially at currently relatively wide spreads.
So my question is: if there is a sectoral shift away from financials, then why is CU / CIU yielding so much?
[…] potential trade was discussed in the comments to a post that posed the question: TCA.PR.X & TCA.PR.Y : What’s Keeping Them […]
[…] two issues were recently highlighted on PrefBlog with the note: There were some credit worries when they made a big investment in Dec 06, but these […]
[…] two issues were recently highlighted on PrefBlog with the note: There were some credit worries when they made a big investment in Dec 06, but these […]