Brookfield Infrastructure has announced:
that it has agreed to issue 5,000,000 Cumulative Class A Preferred Limited Partnership Units, Series 3 (“Series 3 Preferred Units”) on a bought deal basis to a syndicate of underwriters led by RBC Capital Markets, CIBC, Scotiabank and TD Securities Inc. The Series 3 Preferred Units will be issued at a price of $25.00 per unit, for gross proceeds of $125,000,000. Holders of the Series 3 Preferred Units will be entitled to receive a cumulative quarterly fixed distribution at a rate of 5.50% annually for the initial period ending December 31, 2020. Thereafter, the distribution rate will be reset every five years at a rate equal to the greater of: (i) the 5-year Government of Canada bond yield plus 4.53%, and (ii) 5.50%. The Series 3 Preferred Units are redeemable on or after December 31, 2020.
Holders of the Series 3 Preferred Units will have the right, at their option, to reclassify their Series 3 Preferred Units into Cumulative Class A Preferred Limited Partnership Units, Series 4 (“Series 4 Preferred Units”), subject to certain conditions, on December 31, 2020 and on December 31 every 5 years thereafter. Holders of Series 4 Preferred Units will be entitled to receive a cumulative quarterly floating distribution at a rate equal to the 90-day Canadian Treasury Bill yield plus 4.53%.
Brookfield Infrastructure has granted the underwriters an option, exercisable until 48 hours prior to closing, to purchase up to an additional 2,000,000 Series 3 Preferred Units which, if exercised, would increase the gross offering size to $175,000,000. The Series 3 Preferred Units will be offered in all provinces and territories of Canada by way of a supplement to Brookfield Infrastructure’s existing short form base shelf prospectus.
Brookfield Infrastructure intends to use the net proceeds of the issue of the Series 3 Preferred Units for investment opportunities, working capital and other general corporate purposes. The offering of Series 3 Preferred Units is expected to close on or about December 8, 2015.
The sentence “The Series 3 Preferred Units are redeemable on or after December 31, 2020.” is, from what I’ve seen, poorly phrased. My understanding is that it is redeemable at par on every Exchange Date, in line with the accepted structure.
Investors should be aware that the distributions are expected to be a mixture of ordinary income and return of capital for tax purposes; no Eligible Dividends are expected. The company expects a 50-50 split of the two types of income over the next five years, but of course there are no guarantees! I have been supplied with the following characterization of the past five years:
| Past Composition of BIP Distributions | |||||
| 2014 | 2013 | 2012 | 2011 | 2010 | |
| Total distribution | $2.1378 | $1.7883 | $1.4988 | $1.3198 | $1.1277 |
| Total taxable income | $2.1035 | $0.4131 | $0.7939 | $0.4825 | $0.2368 |
| Return of capital | $0.0343 | $1.3752 | $0.7049 | $0.8372 | $0.8909 |
| Income % | 98.40% | 23.10% | 52.97% | 36.56% | 21.00% |
| Return of capital % | 1.60% | 76.90% | 47.03% | 63.44% | 79.00% |
Sure bounces around a lot, doesn’t it?
BIP.PR.A was bid at 20.40 today to yield 5.50% to perpetuity … so call these issues more-or-less even yield. This suggests that the new issue is grossly expensive, unless you place a high value on the “dividend floor” feature, which I don’t.
How do you know there will not be any eligible dividends, from the announcement?
From the announcement, not at all. This conclusion, and the table, is drawn from other sources.
Two questions. First, is there an emerging trend of setting a floor on the reset yield, as this one has done at 5.5%? Second, if there is such a trend and if it continues, do you have any sense how this might affect the preferred share market going forward? My sense is that it should slowly exert a stabilizing influence, but I suspect there might be some unintended/unanticipated consequences as well.
If you can’t determine if there will be eligible dividends when the announcement is released, and these new issues tend to get gobbled up rather fast, how can you make good investment decisions? Do you just guess and hope, or is there somewhere else to look? I assume it would be in the prospectus, but I don’t tend to have access to those until after the new issue is closed.
If you can’t determine if there will be eligible dividends when the announcement is released, and these new issues tend to get gobbled up rather fast, how can you make good investment decisions?
You can’t! Ha-ha! That’s one reason why new issues are difficult to find in portfolios I manage.
First, is there an emerging trend of setting a floor on the reset yield, as this one has done at 5.5%?
Yes. It started with the announcement of CU.PR.I and the feature has been incorporated in every public FixedReset new issue since then.
Second, if there is such a trend and if it continues, do you have any sense how this might affect the preferred share market going forward?
I think the most significant effect is that it’s allowing the issuers to issue new FixedResets in a market that has decided that preferred shares in general and FixedResets in particular are crap.
The market has decided that yields and their dependent dividends have gone down substantially and will therefore continue to go down substantially forever.
Right now, as I’ve stated previously, I don’t think the guarantee is worth much because I don’t think there’s much room for five year yields to fall any more (although eager new issue buyers and their salesmen will be quick to point out that I’ve been saying that for a long time and have been wrong every time).
I think the feature will be a virtually mandatory component of FixedResets going forward, until we’ve had a long period of rising GOC-5 yields, at which point:
(i) A floor rate will be useful and worth something, and
(ii) The market decides that GOC-5 will rise forever and hence floor rates are useless and worthless, and
(iii) Issues will come out with no floor again.
Thanks James. You are probably right that the emergence of floors probably marks the bottom of the fixed-reset market, but it’s also easy for me to see how GOC-5 yields can drop further. If European yields and the massive levels of Canadian and global debt don’t suggest that possibility to you, then you’ll have no interest in the floor. But if you want your preferred shares to trade even remotely like a fixed income instrument, then the floor might be a good thing.
As always, it’s a question of risk-aversion and scenario probabilities going forward. However, as you can see from the calculation of Implied Volatility for the BAM series in the daily market updates, purchasers of BAM.PF.H are paying dearly for the floor.