[leslie]

I am furious with the media's campaign of mis-information regarding the pricing of Canada's Rate-Reset Preferred Shares. Experts have been claiming that these shares drop in value when rates fall. You can see these claims by ScotiaMcLeod, RaymondJames, CIBC, another fund manager, and two interviews on BNN yesterday and today, etc, etc, etc. They are singing from the same song book.

Give me a break. In the first half hour of any introductory course in bonds, everyone learns the mantra "rates down, prices up" and vice versa. When interest rates fall, the value of outstanding debt increases in value. Prices do NOT fall. And rate-reset preferred are no different from other debt. So here is a simple example to prove the point.

Yesterday a new issue PrefA of rate reset preferreds was IPO'd by Company

The 5-yr Treasury rate was 1%
The stipulated spread on the issue was 3%
The coupon for the first 5 years is set at $25 * 4% = $1.00​

Today another new issue PrefB is IPO'd by the same Company

The 5-year Treasury rate dropped in half overnight and is now 0.5%
The stipulated spread on the issue is 3% still because the risk of the company has not changed.
The coupon for the first 5 years is set at $25 * 3.5% = $0.875​

In five years from today your expectations are that ..

The 5-yr Treasury rate will be x%. It is ludicrous to think you can predict any rate different from this x% for reset of PrefA one day previous. So the expected Tbill reset rate will be the same for both issues.
The stipulated spread will be the same 3% for both issues.
The new coupon after the reset of both will be the same $25 * (3% + x%) = $$$​

So back to today. Which would you rather own, PrefA or PrefB?

Both issued by the same company with the same corporate risk.
Both pay the same risk spread over Tbill rates.
Both will be distributing the same $$ coupon at the next reset date in 5 years (give or take a day)
The only difference is that PrefA pays $1.00/yr for 5 years, while PrefB pay only $0.875/yr. PrefA will pay $0.625 MORE in distributions (5yrs * (1.00 - 0.875))..​

The Pref A shares will RISE in value between yesterday and today by the discounted value of their extra dividends.

Rates down, prices up. Just like for every other debt security.

____________________________________________

So what has caused the 20% drop in value of Rate-Reset preferred shares this year?
Well the drop in price has certainly be helped by the campaign of mis-information telling retail investors that their stocks are worth less now because the rates have fallen.
Also, those stocks in the O&G industry have increased in risk a lot this year. So their stipulated spreads are now too low, Yield up, prices down.
But never discount the possibility that the market is actually correctly valued and responding to legitimate factors.

First you have to correctly calculate their effective %yields. The 'experts' may well quote the current yields (= current distribution divided by current stock price). This number is irrelevant and should be TOTALLY IGNORED. If the stock are trading above $25 then they will likely be called at reset, so the normal Yield-to-Call assuming a $25 future value is correct.

For all the other rate-reset you work with today's 5-year Tbill rate, Calculate the coupon as if reset today. Subtract today's actual distribution and discount that difference in 5 years of payments back to today. Adjust the stock price by this discounted value. Divide the reset coupon $ by this adjusted price.

Once you have the effective yield, subtract today's 5-yr Tbill rate to find the market's risk spread. Compare that spread to documented spreads on other debt. There is no public information for Canadian debt, so use the US data. This year's spread has increased from 2.2% to 3.3%.

Using all the Rate-reset issues, except for the O&G industry which are too risky IMO, Ignore all those graded Pref-3 or Pref-4. What you are left with is a list of spreads on Investment-grade rate resets that varies from 1.75% to 3.5%. Right in line with the US data - if not more expensive (too low yields mean too high prices).

For the calculations and data.

 

[Woz]

Depends on the reset date. What you say is true for newly issued/reset preferreds but as you get closer to the reset date, decreases in BOC yields become more significant in the preferred price.

The value of a preferred should be the present value of each of the 5 year terms.

Term 1 (2015-2020): The coupon is fixed during this period. Rates going down increases the value of this term like bonds.
Term 2 (2020-2025): The coupon is not fixed. It depends on BOC 5-year. Rates going down decreases the value as the coupon has been decreased.
Term 3-N: Assume negligible as it’s discounted to present.

For preferreds that have a while to reset, Term 1 is more significant as returns in the 2020-2025 period would have to be discounted to present.

For preferreds that are resetting soon where Term 1 may only last a year. Term 2 is more significant as Term 2 is longer and because it’s only a year away so it’s not reduced as much when discounted to present.

You see this a bit with the current situation. Preferreds that reset soon have been hit harder than the ones that reset in 2018/19. I have started purchasing preferreds that reset in the 2018-2019 as I think they’ve decreased more than they should’ve and now present good value. I think they’re getting swept along with the preferreds that reset soon.

 

[Jets99]

I agree with WOZ, some good value preferreds that reset in the 2018-2019. For example...

Enbridge Series T
Coupon = $1.00 , Trading today at: $15.41 , Effective Yield = 6.5%
Reset date: June 1/19 , Spread = 2.50% , Potential new Yield = 3.3% (If GCAN5YR remains at 0.8%)

 

[leslie]

For preferreds that are resetting soon where Term 1 may only last a year. Term 2 is more significant as Term 2 is longer and because it’s only a year away so it’s not reduced as much when discounted to present.Term 2 (2020-2025): The coupon is not fixed. It depends on BOC 5-year. Rates going down decreases the value as the coupon has been decreased. .

Disagree. Here is the variation on the example above showing why.

4 years ago Company IPO'd PrefA rate reset preferred.

The 5-yr Treasury rate was 1% back then.
The stipulated spread on the issue was 3%
The coupon for the next year (until reset) is set at $25 * 4% = $1.00​

Today Company IPO's another new issue PrefB rate reset

The 5-year Treasury rate dropped in half yesterday and is now 0.5%
The stipulated spread on the issue is 3% still because the risk of the company has not changed.
The coupon for the first 5 years is set at $25 * 3.5% = $0.875​

What is the difference between the issues?
The coupon of PrefA is expected to reset in one year at $0.875 because of today's lower Tbill rate.
For the intervening period, PrefA will pay an extra $0.125 (=1.00 - 0.875).
I would pay more for PrefA than PrefB - equal to the discounted PV of that $0.125.

When rates fall, both PrefA and PrefB are impacted equally. We can only presume that in one year PrefA will reset according to a Tbill rate of 0.5%. The same as today's rate. Both issues will be paying the same $0.875

 

[Jaberwock]

I am sure a lot of the rate re-settable preferreds were bought with the expectation that BOC rates would rise before the rate reset date. That hasn't happened, and many of those issues are seeing a big cut in yield at the reset date, making them less competitive when compared with alternative dividend yielding investments such as REITs and ordinary shares, whose yields have not gone down.

The price drop is not a surprise. It was predictable

 

[leslie]

.... making them less competitive when compared with alternative dividend yielding investments such as REITs and ordinary shares,

I disagree with the idea that debt prices are determined by equity distributions.
If that were so then the T-bill rates would ALSO adjust to common equity distributions and the resulting rate resets would stay high as well.

I do agree that the corporate spreads increase when equity markets tank (making stock dividend yields higher) because of higher risk. But no one is talking about rate resets being priced down because of higher risk premiums - that is what only I am arguing.

 

[Woz]

Your modified example is the snapshot after the treasury rate has decreased. In your example PrefA would not have had a $25 value before or after the rate decrease given the PrefB IPO price.

You need to compare before and after the rate hike to determine the impact to price.

Pre Rate Decrease:
4 years ago Company IPO'd PrefA rate reset preferred.
The 5-yr Treasury rate was 1% back then.
The stipulated spread on the issue was 3%
The coupon for the next year (until reset) is set at $25 * 4% = $1.00

In January Company IPO's another new issue PrefB rate reset
The 5-year Treasury rate was 1% in January
The stipulated spread on the issue is 3% still because the risk of the company has not changed.
The coupon for the first 5 years is set at $25 * 4% = $1.00

They’re both worth $25 pre decrease in treasury rate as people are expecting rates to stay at 1%. The treasury rate then unexpectedly decreases by 0.5%:

Post Rate Decrease:
PrefA rate reset preferred.
The coupon for the next year (until reset) is set at $25 * 4% = $1.00 after it will be $25 * 3.5% = $0.875

PrefB rate reset
The coupon for the next 5 years will be $25 * 4% = $1.00

Pre rate decrease, both were roughly the same value. Post rate decrease, I would rather have PrefB (i.e. PrefA will decrease more in value than PrefB, i.e. short dated resets will decrease more than long dated resets)

 

[leslie]

In your example PrefA would not have had a $25 value before or after the rate decrease given the PrefB IPO price.

Why exactly would PrefA NOT have a $25 price until just after the rate change? You say yourself further down your post that ... "They (PrefA and your different PrefB) are both worth $25 pre decrease in treasury rate as people are expecting rates to stay at 1%."

You need to compare before and after the rate hike to determine the impact to price.

But that is exactly what I did. In my modified example I say ... "I would pay more for PrefA than PrefB - equal to the discounted PV of that $0.125 (=1.00 - 0.875). The price of PrefA should increase by that amount.

I'll restate your example, replacing your PrefB with PrefC because it is different from my PrefB. I'll restate the points in time to agree with my example - the rate change happened last night.

4 years ago Company IPO'd PrefA rate reset preferred.

The 5-yr Treasury rate was 1% back then.
The stipulated spread on the issue was 3%
The coupon for the next year (until reset) is set at $25 * 4% = $1.00​

Yesterday Company IPO's another new issue PrefC rate reset

The 5-year Treasury rate was still the same 1% as 4 years ago
The stipulated spread on the issue is 3% still because the risk of the company has not changed.
The coupon for the first 5 years is set at $25 * 4% = $1.00​

Until now they’re both worth $25.
Last night the treasury rate unexpectedly decreases by 0.5%:

What Happens?

Today Company IPO's another a third issue PrefB rate reset

The 5-year Treasury rate is now 0.5%
The stipulated spread on the issue is 3% still because the risk of the company has not changed.
The coupon for the first 5 years is set at $25 * 3.5% = $0.875​

The price of PrefA increases in price by the discounted excess coupons for the next 1 year. This is the same as my original example. The price goes up $0.125 (=1.00 - 0.875).
The price of PrefC also increases in price by the same discounted excess coupon $0.125 but for 5 years, so it is a bigger increase in price.

Post rate decrease, I would rather have PrefC (i.e. PrefA will decrease more in value than PrefC, i.e. short dated resets will decrease more than long dated resets)

But you have not shown how any of them decreased in price. I show why both A and C increase in price ---- because the 'correct' coupon ($0.875) is represented by B, and both A and C pay more.

 

[doctrine]

Prefs cut in value are a potential long term play; likely to see some hedge funds picking the ones that are trading in the $14-15 range. You see, the one advantage to those prefs is they will not be called by the company. It's like buying a corporate bond at 3.5% that could potentially give you an 80% capital gain in 8 years if BOC rate returned to 3-4% in the 2020s. That is a potentially very attractive return. Not guaranteed, of course, but most investment grade corporates are already yielding in the 3-3.5% range anyway, with zero change of a major capital gain.

 

[Woz]

Hmm, yeah I think you’re right. We’re both in agreement PrefC (long dated) will do better than PrefA (short dated) when rates decrease. I haven’t shown why they would decrease in price and I tend to agree with you now that they shouldn't. It would have to be caused by the credit spread.

 

[leslie]

And what I want is for people to start spreading the word. What the 'expert' say is bunk.
"Rates up, Prices down." "Rates down, prices up"

 

[stantistic]

CPD vs PFF

On a five year chart, why is there such a large divergence in price charts between Canadian CPD and American PFF from 2014 onwards. Don't the Americans have 5 year rate re-sets?

 

[leslie]

Good question. http://stockcharts.com/freecharts/perf.php?CPD.TO,PPS.TO,HFP.TO,HPR.TO,PFF No, they don't have rate resets. But I don't think that fully explains the difference. They traditionally pay a larger coupon - (IMO because they are more risky because they are issued by off-balance-sheet-entities, not the issuing company with all the assets, but I may be wrong about that). The higher distributions plus the higher prices because rates have fallen may explain most. Another reason may be because 2/3 of the holdings they list are trading above par, so investors are mislead by the index's quoted yields which really should be the YieldToCall, which is much lower.

Until recently our Canadian Perpetuals with Investment Grade ratings had a median yield of 5.3% (correctly using YTC for those priced over $25) compared to the PFF's quoted yield 5.3% (not using YTC). That gives a median spread over Long Bonds (Cdn at 2.2% and US at 2.5%) of 3% - which is in line with Corporate Bond spreads. So I think both country's perpetuals are correctly priced. http://www.retailinvestor.org/RateResetPreferredShares.xls

 

[leslie]

I broadcast the challenge to justify the false claims that "RateResets fall in value when interest rates also fall", and got no defensible answers. In total they amounted to these arguments (with my rebuttals) ...
(i) Yes, bonds rise in value when interest rates fall ..... but RateResets are 'special'. Aren't we all.
(ii) They fall because people are disappointed rates did not rise as they expected when originally purchased. They don't want to just suck up the drop in rates like everyone else has had to do, so they make matters worse by driving down the prices so they get a capital loss as well.
(iii) They fall in value because the coupon resets lower, and owners somehow deserve/demand the old higher yields that no other debt holders are now getting.
(iv) They fall in value because they are priced relative to the dividend yields on common shares - while no other debt securities reflect this arbitrage. Heck I don't even value common stock based on the dividend yield.
(v) Their fall in price is validated by companies IPOing new issues at these same high rates. Which does not differentiate the CAUSE of the higher yields (lower Tbill rates or higher corporate rates?).

Prices will fall when the stipulated risk premium is considered too small. This may be industry specific, e.g. the Oil and Gas industry faced higher risk when oil prices were cut in half in 2015. Or, business spreads generally rise when the economy loses steam, as in 2015. Or spreads may rise when the benchmark Treasury changes to higher-yielding Long bonds because prices fall below $25 and the issue become a perpetual. Or spreads may rise because at the original issue, the market was so enchanted with recent capital gains in preferred shares that the issuers were able to IPO at spreads far lower than they should have paid. There are many possible reasons for an increase in the risk spread.

Until the last few days, the average RateReset with an Investment Grade had an effective yield of 3.8% and a spread over 5yrTbills of 3.0% - which is inline with the spread on all other corporate debt http://www.retailinvestor.org/RateResetPreferredShares.xls. So I plump for the last on that list of REASONS for their drop in value. I certainly don't think they are now undervalued unless you believe that the current relatively high 3% corporate spread will drop back. I could argue that the spread may stay high until we see some convincing economic recovery. This last weeks' bounce in Cdn Pref prices may have little support other than a lot of media attention. http://www.theglobeandmail.com/globe-investor/markets/indexes/summary/?q=TXPR-I

 

[leslie]

Two recent articles on the growing risk spread for Corporate debt - the better explanation for the falling RateReset prices:
Buttonwood at http://www.economist.com/news/finan...t-spreads-are-latest-bad-omen-vultures-circle
SeekingAlpha at http://seekingalpha.com/article/358...ets-telling-asset-allocators#comment-62213656

 

[OnlyMyOpinion]

Just to complicate life, we've seen pref prices rising again. CDP is up early 7% from last Wed. Some issues like TD.PF.D are up over 8%.

 

[stantistic]

Rate resets and negative rates

Thanks Leslie for your insightful answers to my post.
Another question I have - the Swiss Central Bank introduced negative interest rates in January, 2015. Suppose that happens in Canada. For preferred shares coming up for reset in this environment, the new rate will be ( BOC rate plus X basis points). If algebraically correct, the new rate for the next 5 years could be utterly disgusting (i.e. less than X basis points). Or am I misunderstanding something ?

http://www.bloomberg.com/news/videos/2015-10-19/the-global-deflation-signaled-by-negative-rates

 

[leslie]

When there is media talk about negative rates you should assume this refers to the overnight rate they control. The premium investors demand for longer maturities would prevent the 5yr rate from going negative unless the world really spirals out of control. But the 5-yr rates would still fall further of course. But if that happens then ALL debt will reflect those smaller yields. New issues will have smaller coupons. Existing issues (of all debt including rateResets) will rise in price until their yields fall enough. As with the point I am trying to make on this thread ...... changing rates does not / will not cause the price of any debt to fall. When rates fall, prices rise. And everyone sucks up the smaller coupon/yields.

The huge bounce back in the prices of rateResets this last week has most probably been driven by the same thing that caused prices to fall ---- the experts talking in the media. First they mislead everyone to think rateReset prices SHOULD fall because of falling interest rates, and last week they started telling everyone that preferred were underpriced (which may, or may not, be true depending on what risk premium you think they should have. It is their risk spread that determines their price changes up and down.

 

[leslie]

Scotia McLeod has a new report on RateReset preferreds that perpetuates this false idea that prices should fall when interest rates also fall. http://www.smlibrary.com/media/docu...ports/Special-Reports/rate_reset_faq_2015.pdf

8. Q: Which yield is more accurate - current yield or yield to reset?
A: Each valuation has its limitations.

Dispute : The correct yield measure to use is NEITHER the current or the ytw. It is the effective yield that factors in today's 5yr rate and the stipulated spread of the preferred, and the preferred's stock price now, adjusted for the higher payments that will be received until the next reset date. Use the calculator at http://www.retailinvestor.org/RateResetPreferredShares.xls

1. Q: Is the drop in price of preferred shares related to deterioration in the credit quality of the companies?
A: It is not attributable to credit deterioration, but due to lower interest rates causing lower dividends following the reset date.

Dispute: There is no way to factually know what causes the market to move. The authors are in no position to discount any particular subjective reason. Bonds are priced according to math with an overlay of assumptions about unknown risks. The math is categorical. Bond prices RISE when rates FALL. The subjective elements to explain why they did move in the opposite direction may have a variety of causes.

The recent IPOs of rateResets with very high spreads for investment grade companies proves that credit quality HAS fallen. Those companies would never have issued the high-spread rateResets if they could have issued normal debt WITHOUT the same high risk premium. The average investment grade rateReset is now (before this last run up in price) trading at a spread of 3% even though they were issued with a 2.2% spread. In the US corporate bond spreads HAVE risen by about 1% this year.

There is also the probability that retail investors have driven down the prices of rateResets BECAUSE articles like this say they SHOULD BE worth less. Or maybe the shares were originally issued at too-low spreads. Or consider that a lot of Oil&Gas industry player have rateResets. We know they are in trouble. Lots of possible reasons.

Only one reason factually cannnot be true - bond prices do not fall because rates fall.

11. Q: How should rateResets trade as they approach the reset date?
A: Preferred shares trade on a yield basis inline with their peers.

Dispute: At last, something I can agree with. Too bad the authors really meant to limit their definition of 'peers' to only other rateResets. They have simply thrown away logic to create a 'special' type of security.

7, Q: What Government bond yields should be used to evaluate the different types of preferred shares?
A: Rate Reset prices should move WITH moves in the 5-yr yield.

Dispute: While the authors agree that perpetual prices move in the opposite direction of rates, for rateResets they throw away the rules learned in their first day of school, and violate the math proof I gave at the top post of this thread.

2. Q: Why are investment grade rateResets down as much as 40% ytd?
A: The rateReset product as a whole has fallen out or favour due to the risk that at their reset date they may be reset with a lower dividend.

Dispute: If yields in the Treasury market fall, so to do the yields in the corporate bond market. So too do yields in the perpetual preferreds market. So too do the yields in the rateReset market. So too for GICs and savings accounts. When rates fall, EVERYONE settles for lower rates. By being willing to sell at lower prices, rateReset owners got hit twice - first they got hit by the falling market rates, and then they accept a too-low BID price and sell for proceeds that won't even cover the cost of buying a replacement debt security paying the now-lower market rate.

 

[andrewf]

leslie,

Do you see any resulting mis-pricing opportunities in ratereset preferreds arising from this confusion/misunderstanding of how to value them wrt changes in interest rates?