PrefBlog

John Heinzl was kind enough to mention me in his Investor Clinic column today titled Why preferred shares plunged, and the ACBs of the Loblaw deal:

James Hymas, who manages a preferred share fund and writes the PrefLetter, said the recent drop may have been exacerbated by tax-loss selling, in which investors dump losing stocks in order to offset capital gains on their winners.

“There’s also a certain amount of ‘fighting the last war’ syndrome,” Mr. Hymas said in an e-mail. “People remember what happened the last couple of times the Bank of Canada cut its [benchmark] rate (January 2015, especially) and are terrified it will happen again.”

Despite those fears, after the recent drop many preferred shares now offer even more attractive yields. That’s especially true considering that preferred shares, unlike corporate bonds, qualify for the dividend tax credit. What’s more, assuming the five-year Canada yield stays roughly where it is, most preferreds will actually raise their dividends – not lower them – on the next reset date, Mr. Hymas said.

“My advice to current holders of preferreds remains the same as it was during the credit crunch: Shut up and clip your coupons,” he said.

Barry Critchley was kind enough to mention me in his piece Assessing the value of the minimum reset yield:

The minimum rate reset feature has some value for the investor and could become real value in five years if the five-year Canada bond rate is below the current five-year rate. The question is whether investors are being rewarded now.

James Hymas, portfolio manager of the Malachite Aggressive Preferred Fund and publisher of the PrefBlog thinks they’re not. “I think it’s overly reflective in the price, but given the current yield on five-year Canada bonds (around 1.70 per cent) the current value is minimal.”

Hymas said the minimum reset has value if in five years, the five-year Canada rate is below 1.70 per cent, a level that’s “significantly below the target inflation rate,” of two per cent set by the Bank of Canada, and not far from the record low five-year rate.

In Hymas’s view, a coupon that starts with a five is required. In the meantime, he says that better value lies in the secondary market.

John Heinzl was kind enough to mention me in his recent article What is a ‘Canada call’ and why was my bond redeemed early?:

In April of 2009, I purchased a 10-year Manulife bond that yields 7.768 per cent and matures on April 8, 2019. However, in early October, I looked at my discount brokerage account and was surprised to see that Manulife had “redeemed” the bond. I have never heard of a bond being redeemed at the behest of the issuer. Is this legal?

…
“That’s known in the business as a Canada call and it’s very common,” said James Hymas, president of Hymas Investment Management. “One of the very important things to do when investing in corporate debt is to look at the call provisions, because they will almost always be there somewhere.”

Generally, companies will only redeem bonds when it is in their best interests (or when required because of the terms of the issue). When Government of Canada bond yields were at historic lows, it wasn’t advantageous for Manulife to redeem the notes because it would have had to pay a steep price. But when government yields started to spike several months ago – and as the maturity date approached – redemption became attractive, Mr Hymas said.
…
Manulife announced on Aug. 15 that it intended to redeem the notes and, on Oct. 3 it said the redemption price would be $1,073.81 (per $1,000 face amount) plus accrued interest of $38.52. The redemption price was based on the second option in the prospectus, as it was higher than par. This equates to a yield to maturity (YTM) of 2.73 per cent, Mr. Hymas said. (If you’re wondering why the YTM is lower than the 7.768-per-cent coupon rate, it’s because the notes were trading above par.)
…
“They essentially bought back their old debt at a yield of 2.73 per cent and were able to replace that with an extension of term of more than five years and with debt that was actually subordinated and that’s a good deal for them. They’re only paying about 32 basis points [in additional yield] and that’s a bargain,” he said. (Subordinated debt, with its higher risk from a bondholder’s perspective, would normally carry a higher interest rate than senior debt.)

Rob Carrick was kind enough to quote me in his piece Are these preferreds a 5-per-cent solution?:

James Hymas, president of Hymas Investment Management and a preferred-share specialist, said there are roughly a dozen blue-chip perpetual issues that combine a straightforward structure with a share price below their $25 issue price. These shares, issued by Brookfield Asset Management, Canadian Utilities, E-L Financial, Power Corp. of Canada and Power Financial, had an average yield of 5.1 per cent as of earlier this week.

Mr. Hymas said perpetuals could be an attractive option for investors who are looking at long-term corporate bonds, which mature in 10 years or more. These bonds have a yield of roughly 3.6 per cent on average today compared with perpetuals in the range of 5 per cent.

You’d need about 6.6 per cent from a bond to give you the same after-tax yield as a perpetual preferred share. Multiply a dividend yield by 1.3 to get the rough equivalent from a bond after paying taxes, Mr. Hymas said.

The drawback with perpetuals is that they are highly sensitive to rising interest rates. Mr. Hymas sees five-year Government of Canada bonds eventually rising from current levels just above 1 per cent to the 3-per-cent to 3.5-per-cent range. If that happens, expect perpetuals to plunge in price. “It could be quite traumatic,” he said.

Much less of a concern is the security of the dividends paid by perpetuals. The dozen perpetuals highlighted by Mr. Hymas are all rated Pfd-2 by the rating agency DBRS, which means satisfactory credit quality. Pfd-1 is superior quality, while Pfd-5 is highly speculative.

Companies that issue perpetuals have an option to redeem the shares based on terms set out in the prospectus (you can find prospectuses on Sedar.com). While you shouldn’t buy these shares based on the potential for a future redemption at the $25 issue price, there are some perpetuals that might actually be in line for this at some point in the next several years.

Mr. Hymas believes that regulations covering the banking sector and its capital structure could, in the next few years, be applied to life insurance companies. Some preferred shares would then become disadvantageous for insurers and, in turn, be candidates for redemption in the years ahead. A redemption would turn a perpetual preferred share into a maturing investment similar to a bond.

Mr. Hymas said he likes three issues of insurance-company perpetuals with redemption potential: Sun Life Financial Series 5 (SLF.PR.E), Manulife Financial Series 3 (MFC.PR.C) and Great-West Lifeco Series I (GWO.PR.I). All have yields at or just below 5 per cent based on recent share prices and dividends. If the shares were redeemed in the next eight to 12 years at $25, that yield would work out to be a bit higher (the shares all currently trade below $25).

For argument regarding my belief “that regulations covering the banking sector and its capital structure could, in the next few years, be applied to life insurance companies”, see the posts DeemedRetractible Review: September, 2016 and OSFI Dovish on Insurance Tier 1 Eligibility Rule.

Barry Critchley was kind enough to quote me in his article TransAlta plays the Grinch with its preferred share holders:

James Hymas, who runs Hymas Investment Management and who also publishes the PrefBlog, wrote “all of this analysis leads to the conclusion that this is a rotten deal for the preferred shareholders, so rotten that we may call it a sleazy attempt by the company to pull the wool over the eyes of unsophisticated retail investors. As the company admits, they look forward to reducing the corporation’s notional capital balance of preferred shares by approximately $300 million.”

After noting that an analysis based on implied volatility would require an even higher dividend than the 6.50 per cent TransAlta is offering, Hymas said the $300-million “is money that currently can potentially be earned by the current shareholders.”

That $300 million could occur with price increases on the extant issues; from an increase in the five-year government of Canada yield, or “from straightforward spread narrowing. The company is giving up nothing – NOTHING! – in order to capture this entire amount for themselves,” he wrote.

Assiduous Readers will remember that my views on the proposed Exchange (which will be voted on as a Plan of Arrangement) were published in the post TA Proposes Sleazy Exchange Offer.

Update, 2016-12-24 I was perplexed by a comment on Financial Wisdom Forum:

More on the TransAlta exchange.

http://business.financialpost.com/news/ … picks=true

FWIW, I am quite satisfied with the offer because I’m a trader and am more than happy to bail on these PF-3 issues because I really believe that one would have to be wearing super sized rose coloured glasses to think that they would someday trade or be redeemed at par, especially with a company like TA that has slashed the dividend on the common to 4 cents/quarter.

The case for the “No” vote does not depend on the hope that the shares will “someday trade or be redeemed at par”, and demonstrating this should actually make the argument more clear for those who have difficulty with the concept of Implied Volatility.

Let us examine the specific case of TA.PR.D; the following analysis framework may be applied to the other series with changes in numbers.

TA.PR.D:

• pays $0.67725 p.a. until the next Exchange Date
• will reset to GOC-5 + 203bp (paid on par value of $25) on each Exchange Date
  • This is equal to (25 * GOC-5) + (25 * 203bp)
  • which is equal to (25 * GOC-5) + $0.5075
• may be redeemed at $25 on each Exchange Date
• Exchange Dates are 2021-3-31 and every five years thereafter

The company proposes to exchange each share of this for 0.503 of a New Preferred Share; each New Preferred Share will

• Pay 6.50% of $25.00 = 1.625 until the next Exchange Date
• will reset to GOC-5 + 529bp (paid on par value of $25) on each Exchange Date
• may be redeemed at $25 on each Exchange Date
• Exchange Dates are 2021-12-31 and every five years thereafter

The fact that holders will be getting only 0.503 New Preferred Shares for each share of TA.PR.D makes the changes a little more complex for many investors, so as a thought experiment, let’s design a Notional Share which we will assume will be offered 1 for 1 for TA.PR.D, with the new holdings, in total, having exactly the same characteristics as the proposed new holdings of the New Preferred Shares.

A Notional Preferred Share:

• pays $0.817375 until the next Exchange Date
• will reset to 0.503 (GOC-5 + 529bp) * 25 on each Exchange Date
  • This is equal to (0.503 * 25 * GOC-5) + (0.503 * 25 * 529bp)
  • which is equal to 12.575 * GOC-5 + $0.6652175
  • subject to a minimum rate of $0.817375
• may be redeemed at $12.575 on each Exchange Date
• Exchange Dates are 2021-12-31 and every five years thereafter

So when we compare the currently held TA.PR.D to the Notional Share we see that:

• The Notional Share will pay an extra $0.14 annually for each of the next five years (approximately), for a total of $0.70.
• The redemption price will drop from $25 to $12.575
• The dividends after the next Exchange Date (if it is left outstanding) will depend on the GOC-5 yield, as indicated on the following chart

Click for Big

The big problem, of course, is the change in redemption price – holders lose out on a lot of potential capital gains if the market improves, either through increases in the GOC-5 yield (which should increase the trading price of the preferreds) or through a narrowing of spreads (which may occur because the market improves, or TA’s credit improves, or both). In addition, we see that increases in the GOC-5 rate greatly improve the dividend payout from TA.PR.D and the much higher redemption price means these potential increases will not be called away unless for a gigantic premium over the current price.

John Heinzl was kind enough to quote me in his piece This borrowing-to-invest strategy is no slam dunk, which discusses a couple’s proposal to take out a home-equity loan and invest the proceeds in ZPR, the BMO Laddered Preferred Share Index ETF:

If you look up ZPR’s distribution history, however, you’ll see that the monthly payout has been dropping. At the start of 2014 it was 5.3 cents, a year later it was 4.8 cents and currently it’s 4.5 cents. The reason: ZPR holds a basket of rate-reset preferred shares, many of which have cut their dividends because of a steep drop in the five-year Government of Canada bond yield that is used to set the dividends on these shares.

Barring a rebound in bond yields, ZPR’s distribution could fall even further as more rate-reset preferred shares cut their dividends, says James Hymas, president of Hymas Investment Management. Assuming the five-year Canada yield remains at its current level of about 0.8 per cent, the “implied current yield” of ZPR’s underlying portfolio is just 4.74 per cent, according to a BMO document dated April 29. After deducting the ETF’s management expense ratio of 0.5 per cent, ZPR’s net implied yield to investors is about 4.24 per cent, Mr. Hymas says.

That’s not nearly as attractive as the 5.6-per-cent yield for ZPR advertised on BMO’s website.

The document mentioned in the article is available here … for now! Hat tip to MikeFreedom49? on the Financial Wisdom Forum for bringing this document to my attention.

This is probably a good place to make two other points.

First, the document states that the duration of ZPR is 3.11 years, with a footnote indicating that:

Duration is a measure of sensitivity to changes in interest rates. For example, a 5 year duration means the value will decrease by 5% if interest rates rise 1% and increase in value by 5% if interest rates fall 1%. Generally, the higher the duration the more volatile the price will be when interest rates change.

This assertion of a 3.11 year duration without any qualifiers is irresponsible, reckless and wrong. Such an answer in the current environment can only be derived by pegging the price on the next reset date at an arbitrary level – it could be the current price, it could be $25.00, it could be anything you like, as long as it’s presumed to be a value unaffected by market yields. These are perpetual instruments trading below par; their duration in the current environment is better in the 20 to 25 year range (with respect to spreads) … or it could even be negative (if we assume yield-to-perpetuity will be constant in the face of a changing 5-year Canada rate)! . Holders of ZPR – and other FixedResets – over the past two years will doubtless be happy to confirm that the price volatility they have experienced far exceeds what they might have expected from an investment with a 3.11 year duration.

Classical bond mathematics is a very useful tool for examining preferred shares, but must be accompanied by a very cautious and explicit statement of assumptions and these assumptions should be disclosed.

Secondly, I note that BMO touts the fund as being “Low to Medium Risk”, which I feel confident will be considered laughable by those who have suffered through the woes of the preferred share market for the past two years.

The whole concept of these risk assessments, with their mandatory reporting on the “Fund Facts” statements is under constant attack by those with even a smidgen of knowledge about risk (which leaves out the regulators), so I won’t go into this further except to say I’d love to see how they justify their claim!

Finally, here’s a chart of expected changes in the dividends of Fixed Resets for which the YTW scenario is perpetuity, taken from the May, 2016, edition of PrefLetter:

Click for Big