ACO.PR.A: Negative Yield-to-Worst; Positive Yield-to-Issuer-Best

I recently had a contest about ACO.PR.A which has been won by Assiduous Reader adrian2.

The question was:

So: here’s the question … how might a rational investor reason that paying $27.00 for this issue has enough chance of at least a half-way decent return to make it worth while? This investor knows that the yield to worst is negative and that he’s taking a chance … why might he buy it anyway?

ACO.PR.A closed at 26.62-75 on Friday, so the situation isn’t quite as dramatic as it was when the bid was $27.00. However, the portfolio of possibilities for this issue on Friday, according to HIMIPref™ was:

Call 2008-12-31 YTM: -1.78 % [Restricted: -0.54 %] (Prob: 40.10 %)
Call 2009-12-31 YTM: 2.35 % [Restricted: 2.35 %] (Prob: 1.01 %)
Soft Maturity 2011-11-30 YTM: 3.68 % [Restricted: 3.68 %] (Prob: 58.89 %)

So the YTW was -1.78%.

Why would an investor pay such a rich price for the issue? Well, there was a hint of the answer buried in my article about retractible preferreds and bonds:

One may sometimes make a reasonable argument that YTW is not the most appropriate method of calculating yields. Say, for instance, that a company has the ability to issue preferreds that pay $1.25 p.a. and has an issue outstanding that pays $1.40 and is currently callable at $26.00, with this price declining by $0.25 annually for the next four years. If the issue is trading above $26.00, the YTW scenario will almost certainly be an immediate call. However, since the company can save $0.25 by delaying redemption, the net cost to the company of leaving the shares outstanding for another year is the dividend less the saving, or $1.15 (during the declining call-price period). Since this is less than the rate it would pay on a new issue, the company may well prefer to wait.

The question of what to believe is a complex question—complex enough, in fact, that I am currently devoting a great deal of time to researching the matter. Most investors will be well advised to rely on YTW.

In other words, Atco (the issuer of ACO.PR.A) may look at the issue on 2008-12-1 and say – ‘well, this issue pays $1.4375 annually, or 5.75% of par (as dividends, which costs more than interest) … we’d better redeem.’

But according to the redemption schedule, they can save $0.50 off the redemption price every year by waiting! If we were going to analyze this precisely, we would look at the situation from their point of view as a loan of $26.00 which amortizes down to a $25.00 balloon payment on 2010-12-1 (the first date they can redeem at $25.00) via quarterly blended payments that include principal. But from a back-of-an-envelope, conceptual perspective, we can say that it’s a loan of $25.50 with an cost of $1.4375 dividend LESS the decrease of $0.50 in redemption price, for an all-in cost of $0.9375 … which comes to about 3.67%, a financing cost calculated to bring smiles to the most hardened CFO, even if it as expensive dividends.

These calculations will be incorporated into the next release of HIMIPref™ – yes, I am working on it, albeit slowly – as “Yield to Issuer Best”. To select YTIB from the table of possibilities, I will be determining the disparity of each redemption option from the yield curve – that is, cash flows for the instrument will be discounted by the self-consistent yield curve to arrive at a net present value for each option. The lowest NPV will determine the YTIB. I do not know, at this point, how influential in the valuation YTIB will be relative to YTW … but hey, it’s worth looking at!

Incidentally, ACO.PR.A was added to the TXPR index in the last rebalancing, despite what might be considered to be relatively low average trading value (HIMIPref™ indicates $18,374 worth of shares daily, TMXMoney calculated 1,600 shares (the difference doesn’t bother me; HIMIPref™ is an adjusted exponential moving average; I believe that TMXMoney is an unadjusted rolling average, virtually guaranteed to show a higher result). A large part of ACO.PR.A’s current price may well be buying from indexers.

The following graphs regarding ACO.PR.A have been prepared and uploaded:

3 Responses to “ACO.PR.A: Negative Yield-to-Worst; Positive Yield-to-Issuer-Best”

  1. kaspu says:

    The best thing about this whole credit fiasco:

    I get the biggest “I told You So” in my 26 year marriage for all the times my wife yelled at me for concentrating on prefs rather than Google.

  2. […] some time ago when I had a contest about the analysis of issues with a negative yield-to-worst. The explanation referred to ACO.PR.A, but it can be applied to […]

  3. […] was last mentioned on PrefBlog in connection with the calculation of yield-to-issuer-best. It is tracked by HIMIPref™ and is a constituent of the Operating Retractible […]

Leave a Reply

You must be logged in to post a comment.