Part of my being wrong must be attributable to superficial research. I have to be able to dig deeper into issues, and I shouldn't just arbitrarily conclude on issues without looking at every angle. Perhaps this is a good goal to establish for 2016.
I began looking at these bonds again recently because I really can't get my head around why they are trading on par with other equivalent speculative grade issuers.
Here's the TD listing today. These appear to be the highest rated non-investment grade bonds in TD's inventory apart from Great Cdn Gaming maturing in 2022, but with a +237.5 BPS better coupon.
And here's Transalta, Nov 2020, 5% coupons, maturing 7 months after the Corus issue, and with a higher credit rating to boot! BBB / Stable vs BBH for Corus. The fact that Transalta corporates with a 75 BPS better coupon and 7 months longer duration are trading -1.70 lower than Corus corporates rated lower tells me that the market may not think much of DBRS' BBB Stable rating (unless the differential is entirely due to 7 months duration, but I don't think so).
Here's Transalta (common) over the last 5 years:
And here's Corus (common) over the same period:
If one were superficial enough, they would stop at the price charts only and conclude that both companies were doomed by virtue of what appears to be irreparable damage to price with equal velocity. Ok, they have almost the same slope throughout 2015, but I think the comparisons should stop there.
I examined Corus' last 8 years of financial statements and here's what I found:
- In every year since 2008, interest coverage has been greater than 4.5x with the exception of 2010 when it was 4.21x.
- On a free cash flow basis, the business (as bad as it is perceived to be), produces free cash flow. This free cash flow has been used to pay generous monthly dividends (I have no idea whether or not they will cut the dividend, perhaps the decline in the stock price already reflects a dividend cut)
- I think that the pertinent question is whether the new CRTC pick and pay rules will have a material adverse impact on Corus' ability to generate the same level of subscriber fees. Will the impact of pick and pay end up chopping operating income of $234M in 1/2 next year? Or will something else happen. I stumbled across the following article on Seeking Alpha discussing this exact issue, here are a few quotes from the article, link here:
So the author's take seems to be that pick and pay may actually help Corus, contrary to conventional thinking and certainly contrary to what most BNN interviewees (pros) are rambling on about, as follows:
So my first exercise was to revisit my initial assumptions regarding bankruptcy outlined in my previous post. If memory serves me correctly, I analyzed the balance sheet on a liquidation basis under three scenarios and concluded that the risk wasn't worth the reward on an equal probability weighted basis in terms of considering the bonds for purchase.
Here's where I think was wrong:
I assigned equal weight probabilities to liquidation scenarios without considering whether liquidation was actually a legitimate outcome to consider. The company has taken steps recently to expand content offerings (via deals with Disney, Nickelodeon), and has divested its pay TV business for $200M+ in a deal with Bell, raising cash to reduce debt.
I still think I have to consider a chance of liquidation though, but I don't think it's as much as 1/3 as per my original scenario.
Here's the balance sheet at Aug 31, 2015
So here's how I revisited the analysis, assuming I bought the bonds at 94.50, maturing in 2020:
- Assume worst case is liquidation. In this case, the current assets (cash, A/R and tax recoverable), get pretty close to 100 cents on the $1. For the long term tax credits, I'm going to assume the same. The film investments and the licenses - maybe $.50 on the $1.
- Liabilities paid first are the secured creditors, the bank loan facilities. After that, the unsecured creditors get the liquidation proceeds according to who yells the loudest. Unfortunately, the 2020 bonds are unsecured obligations so these rank alongside other unsecured creditors.
- I calculated that in liquidation, the bondholders would likely get around $.80 on the $1 (but likely less if I haven't considered off balance sheet obligations such as premise leases)
What's the probability of liquidation? No one knows. If pick and pay does not adversely affect the company in the way expected, then the probability is likely low. Let's say between 10% and 20%, and select the mid-point, or 15%.
My loss on the bonds would then be:
($.80 - $.945) / $.945 = -15% (not counting interest paid up to liquidation) x 15% probability = -2.25%
If the company does not liquidate and I get made whole in 2020, my return on the bonds is:
($1.00/$.945) ^ (1/4) - 1 + YTM of 5.77% = 7.20% x 85% probability = 6.12%
So my reward risk looks to be 6.12% : 2.25% or almost 3;1
I'll probably get a better idea of how the pick and pay impact unfolds as we get closer to the end of 2016 early 2017, but I'm willing to bet against the consensus opinion now that pick and pay will be the death knell and as a result, I bought $1,000 face value of the bonds today at 94.6. I didn't feel comfortable with $5,000, which is the typical minimum for trading investment grade bonds at TD, but the bond desk must have either thought it was my lucky day, or they must have felt sorry for a small fish and shown me some charity (or worse - having read Liar's Poker).
There's one more point to consider: a takeout. On a takeout / change in control / acquisition of Corus, the bond indenture includes a provision to allow the holders to be made whole at 101% plus accrued interest. That would certainly be nice. The sooner it happens (if it were to happen at all), the juicier the return gets:
After year 1, return = (101/94.60) - 1 +5.77% = 12.5%
After year 2, return = (101/94.60)^(1/2) - 1 +5.77% = 9.1%