Yellow Media: Cigar Butt Arbitrage

Note: I wrote this up before putting on the trade last week. While the shares have gone up a little the thesis remains the same.

Yellow Media (TSE:YLO) is a small Canadian company that is recapitalizing. YLO hopes to exchange $1.8B worth of debt for new debt, cash, and 82.5% of New Common Shares (NCS). The existing convertible debentures, preferred shares and common shares will be exchanged for 17.5% of NCS and warrants (representing in the aggregate 10% of NCS). They have holders of 30.0% of medium term notes and 23.7% of Senior Unsecured Debt supporting the Recap.

If this goes through the company will be significantly less levered (Net debt of $794M) and save approximately $45M a year in interest alone.

Conversion Rates:

From the Q&A document common shares will be treated as follows:

"Each holder of 100 existing common shares will receive 0.50000 New Common Shares and 0.28571 Warrants that are exchangeable into New Common Shares."

So right now every 100 common shares that are bought will cost $7 and in exchange you'll get 1/2 share and 0.28571 warrant which will exercise at $31.67 (and expire 10 years from the closing of the Recap). Ignoring the value of the warrants, NCS are valued at $14 right now.

The company also had existing preferred shares. Preferred shares 1,2,3 and 5 will receive the following:
"Each holder of 100 existing preferred shares, other than Series 7 preferred shares, will receive 6.25000 New Common Shares and 3.57143 Warrants that are exchangeable into New Common Shares." Cumulative dividends on the existing preferred shares won't be paid."

Today the series 3 (TSE: YLO.C) will cost you $53 to buy 100 shares. After the recap you will receive 6.25 NCS and 3.57143 warrants. YLO.C is pricing the NCS at $8.48 a share($53/6.25), assuming that the warrants are worthless.

Game Theory

There are two scenarios that could play out in my eyes. The recapitalization goes through (1) or it doesn't(2).
(1) If the recap goes through the logical argument is to go long the preferred. Right now the shares of the preferred are priced at a 39% discount to the existing common shares. While it is unknown whether or not the deal will go through it seems likely in my opinion.

As stated previously, management has commitments from note holders of the medium term notes (30%) and Senior Unsecured Debt (23.7%). If this does go through a holder of $1,000 face value medium term notes will get Senior Secured notes($423), subordinated Unsecured Exchangeable Debentures($56 exchangeable at a conversion price of $21.95), NCS (12.001 shares) and cash ($141). The new Senior Secured notes will pay 9% and will be redeemed with 70% of consolidated excess cash flow at par on a pro rata basis. Ignoring the value of the common shares, existing bond holders are getting 62 cents on the dollar.

(2) If the recap doesn't go through it's a whole new ballgame. The worst case scenario is they file for bankruptcy and the common goes to zero. I have no ability to predict what residual value (if any) will exist in the common or preferred so I won't try. The preferred, being higher up in the capital structure, will have more residual value.

So the question is: how do you play the upside potential that this recap has while drastically reducing the downside of a bankruptcy filing?

Go long the preferred and short the common.

While the upside is less, the downside is also capped. By shorting the common you negate the bankruptcy risks and by going long the preferred you get a play on the upside if the recap goes through. The image below shows what we believe will happen if the recap goes through, you buy a 1000 share slug in the preferred and short an equal dollar amount of common.(This has changed obviously due to price fluctuations)

The delta is the number of shares you would receive post recap. While the value of the NCS is not yet known, the imagination can run wild(see next section). By holding the preferred one also gets 35.7 warrants. The value of the warrants has been completely ignored for now.

The downside is the borrowing costs (~3% annual borrow cost currently) and transaction costs. It is hard to imagine a scenario where the preferred get totally screwed but the common survive. Currently convertible debentures holders are crying foul over the recap but they seem to be the only ones left in the dark.

"However, during a conference call on Monday, Yellow Media’s restructuring advisers, BMO Nesbitt Burns and Canaccord Genuity, explained the rationale for treating convertible holders in this manner.

They said that Yellow Media assumed that convertible debenture holders would convert their bonds to common shares, despite an $8 conversion price when the stock was trading around eight cents.

According to lawyers not involved in the deal, the presiding court for this restructuring will likely rely on case law to approve the deal, and case law dictates that debt holders and shareholders must vote on the transaction as separate classes of owners. To make sure shareholders approve, the company must give them reason to vote, and that could be why they’re getting something extra.

Convertible debenture holders, however, have been grouped with holders of the senior debt. Their $200-million pales in comparison to the $1.8-billion owed to senior holders and the banks, so their votes are not going to sway the result."

The convertible holders appear to be retail holders. A significant portion of senior holders have already agreed to the restructuring. Holders who appear to be more sophisticated investors (Bloomberg screen shots). To sum it up...I don't see how holders of convertibles can sway the vote enough.  

What could the NCS be worth?

I believe the downside from this trade is defined. If the recap does not go through I would expect the common to drop for fear of bankruptcy and the preferred to drop but less than the common given the discount already present and the higher residual value in bankruptcy. I anticipate a mark to market downside could be much greater than stated but patience should take care of that. 

For fun, let's assume that the recap goes through. What is a reasonable price for the NCS?

Net debt/LTM EBITDA will be 1.3X, Net Debt is $794M which means EBITDA is $610M. LTM cash interest was $132.7M and taxes were $134.8M(YLO financial filings). The press release for the recap stated that interest expense would be reduced by $45M. So post recap earnings before CapEx are $387.5M. For the past four years CapEx has averaged $52.4M. I'll round up to $60M, so earnings would be $327.5M. Since YLO is a dying business lets say that it's only $300M.

There are going to be 26M shares after the recap plus 3M warrants which are ITM at $31.67. If the NCS are priced 2X earnings or $600M, with 26M shares that's $23/share. Above $21.97/share the Unsecured Exchange Debentures are ITM.

Conclusion:

By going long the preferred and short the common of YLO one is able to significantly reduce the risk present in this recapitalization. Even if bankruptcy is declared a low cost short can be executed, negating most of the downside risk.

On the other side of this arbitrage there appears to be plenty of upside if the recap goes through. Ultimately it is hoped the recap goes through, the shares shorted are covered and the owner of preferred has a significant stake in a company that has little debt and significant upside, even in a business that is dying. Long YLO.PR.C, Short YLO. 

 

Update 9/11/2012

 

The recapitalization went through. I covered 3/4 of my short leaving only a little remaining in case the court order goes against them. I didn't really think the change in the recap made much of a difference. It basically dilutes NCS by ~8%. Nothing really significant.

 

Thus far this has been a profitable trade as I've been able to take advantage of an arbitrage several times. Still long YLO.PR.C and short YLO.

A CONN-fusing company

Business:
Conn's is a retailer. They sell electronics, furniture,  and appliances to the thousands of sub-prime borrowers who walk into their store. About 80% of their sales are financed.

Goal of the analysis: Determine if Conn's is overvalued and a timely short

Amazon is decimating brick and mortar retailers. As expected, the stocks of Best Buy (BBY) and hhgregg (HGG) have been punished and appear to be value traps. I believe that Conn's should be valued similarly, if not worse than those two companies. While analysts will argue that the company is growing quicker than rivals and has an advantage by extending credit to those who can't afford it I think there are problems.

Is CONN growing enough to deserve a premium?

The charts below show revenue and store growth is non-existent at CONN. HGG and BBY have been growing store count and along with that, overall revenue. Conn's store count has dropped from 76 in 2009 down to 65. It does seem that these were less active stores as revenue/store has risen from $9.59M in 2010 to $10.05M in 2012(~2% per year).  

HHG is argued to be snapping up old Circuit City stores. This rumor seems believable based on their store count numbers quickly expanding since 2008. BBY is growing stores and revenue is starting to flatten out, in their 10K they mention that comparable store sales have declined by ~1.7% per year the past two years.

Regardless, CONN is the only one of the bunch that is neither growing store count nor revenue. While they have eliminated stores that are less active (based on Rev/store number) the growth within existing stores is hardly inspiring. Yet this growth commands a 13X forward multiple based on guidance of $1.30 EPS.

This high multiple for a sector in decline seems odd when you consider BBY has a forward multiple of ~6X with guidance for GAAP EPS of $2.85-$3.25. HHG is guiding for GAAP EPS of $0.90-$1.05 for a multiple of ~6.6X.  So why would CONN be valued 2X as high as rivals who may be in a secular decline, but are still quite profitable, growing revenue and store count? 

Does Conn's have a better business model?

Maybe CONN is doing something better than their rivals. Perhaps they sell different products, don't have as many contractual obligations, or get much higher returns through their financing arm.

Besides the subprime financing, BBY and HHG sell cell phones while CONN does not. In FY2012 BBY had 40% of their sales come from "Computing and Mobile phones." Besides appliances (comparable store growth up 10.6% from 2012 to 2011) it's been the only growing segment up 6.0% compared to 2011 in comparable stores.

For HHG 9% of sales came from computing and mobile phones and comparable store sales grew 52.9% YoY. Appliances grew 3.0% YoY in comparable stores. Besides that, all other segments declined but they did mention that mattresses had "strong double digit growth.''(pg 37 2012 10K)

So appliances are growing between 3-10% at the two other retailers. At CONN recent conference calls have emphasized increased furniture and mattress sales. This segment increased 35.2% YoY (pg 8 2012 10K) but only makes up 16.7% of their product sales. The thought is that they can generate higher margins by growing this segment. Retail gross margin has improved 2.2% YoY thanks to an emphasis on furniture and mattresses. I think the obvious thing is that cell phones should not be focused on(see Radioshack).

So there are some differences in the business model. So much in fact, that for 2013 management expects retail gross margin to jump up to 33%, largely thanks to product mix. I still believe that's a lofty goal considering a well established company like La-Z-Boy has 30% gross margins currently and 10 year gross margin average of 26.5%(LZB filings).

Total contractual obligations are always big for retailers as operating leases can be hidden. BBY has total contractual liabilities of $14.2B (pg 54 2012 10K) for a total capitalization (MC+total contractual liabilities-cash) of $19.1B. With a low end earnings of $2.85 and 366.3M total shares net income will be $1.043B. So BBY has a total capitalization to earnings ratio of 18.3X.

HHG has $715M of contractual obligations, no debt and cash of $59M. With a market cap of $227M they have a total capitalization of roughly $880M. Net income will come in at $36M if EPS is $0.95/share. So my calculated total capitalization to earnings ratio for HHG is 24X.

Finally, CONN has $458M of total contractual cash obligations (pg 59 2012 10K), a market cap of $578M and cash of $6M (which isn't excess but I'll just subtract it for simplicity) for a total capitalization of $1030M. With net income of $1.30/share and 32.9M diluted shares (a 1M increase from the quarter before!), my calculated total capitalization to expected earnings ratio is 24.1X.

I could calculate ratio after ratio and paint a million different pictures. At the very least I'm convinced that HHG isn't that cheap due to contractual obligations and neither is CONN. Yet all three are in the same ballpark. What I can try to determine now is the quality of Conn's earnings. Are they avoiding the secular decline in brick and mortar stores from their financing arm?

Conn's Financing and Overall Valuation

This credit segment first caught my eye and still mystifies me. Somehow this company charges 18% interest rates and has net charge-offs of 8.5% on a $634M portfolio balance(pg 23 Q1 2013 10Q). Therefore the maximum amount of money they'll get is 18% of $634M ($114.1M). They also have 3rd party financing that makes ~$25M. Total financing revenue is $138.6M. 

If we back out the expected 8.5% net charge-offs ($53.89M), $24.5M of interest expense (guided to on their Q1 2013 earnings call) and Credit Segment SG&A of $57.8M (pg 42 10K 2012 41.7% of revenues), we have total costs for the portfolio of $136.2M. Adding back the provision for bad debts of $53.6M and the $634M portfolio makes $56M or 8.8%. 

What I find amazing, stunning actually, is that their loss rates are so low. They re-age their loans and currently 11.6% of their loans are re-aged. CONN has figured out something that Visa, MasterCard, Bank of America, Citi, JP Morgan, etc have not figured out. Giving a group of people with an average weighted score of 601 more chances at paying back debt drops charge offs to rates seen in prime borrowers.



According to Lending Club's white paper, FICO bands of 600 or lower will see >19% charge offs. Hmmm, if we add 11.6% plus 8.5% we get 20.1%. I would argue that re-aging simply pushes out the charge-off date. I believe this is confirmed even with secured debt. On home loans with LTV's over 90% (the average down payment for Conn's is 4.5%) foreclosure rates average 25% for FICO 580-599 and 19.5% for 600-619. It's simply amazing that CONN has such low rates of default and in reality, they probably don't.

FY 2012 Valuation

So let's revisit the business again and see if it's making money. The information below is from the 2012 10K and simply reproducing their numbers to fit what I see as free cash flow.

The retail section is pretty straightforward. Retail gross profit for FY 2012 was $191.6M ($653.6M-$455.5M-$6.5M). Total SG&A was $237.9M(I'll assume all SG&A is in retail, by my calculations Credit SG&A minus provision for losses is only about $3-4M). Therefore operating profit for this segment was -$46.3M.

Credit had total revenue of $138.6M, net charge offs of $46.9M, and borrowing costs of $22.5M. Profit for the credit segment was $69.2M.

Other non-operating charges included $7M for store closings and $2M for impairment. These are added back to operating cash flow on the cash flow statement but I believe that to be a mistake. Opening and closing stores, even if they are not operating, still cost a business money. Leases have to be paid and that payment is usually in cash, meaning it's a real cost.

From there I've assumed that a real investor would see D&A ($12.8M), stock comp ($2.3M), and loss from early extinguishment of debt ($11.0M) for total non-cash provisions of $26.1M. Back out $4.3M in CapEx and the company generated $35.7M in what I believe is free cash flow.

With today's market cap right around $578M the investor today gets a FCF yield of 6.1%.

FY 2013 Valuation

But perhaps next year will get better. The stock market is after all, future looking. They hope to open 5-7 more stores in FY 2013. Let's say 6 stores are opened and for simplicity sake, they make 1/2 the revenue that established stores make (50% of 10M).

In FY 2013 if sales grow by 2%(what I've calculated rev/store growth to be over 2 years) and an extra $30M is made 2013 revenues will be $696M. Maybe they become a furniture store for people with credit scores under 605, if so they could see 30% gross profit (in line with La-Z-Boy). So gross profit would be $208.8M. SG&A should rise ~5% due to 10% more stores for half a year so SG&A of $249.8M. So operating profit should come in around -$41M.

The financing segment is tough to predict but lets assume that it too increases by ~10% and the average total outstanding balance gets to $700M, roughly what it was for FY 2011. At 18% interest they'll make $126M plus another $28M (a 10% increase from 2012) from third parties. Total Financing revenue of $154M. Subtracting out 7.5% net charge-offs (7.5%*$700=$52.5M), $24.5M interest (guided on their Q1 2013 call) the financing segment should see operating profit of $77M.

If we add back non-cash of $26.1M to the two segments operating profit of $36M ($77M-$41M) and back out CapEx for the year of $18M (pg 58 10K 2012) FCF for the year is $44.1M. So forward looking an investor today can get a business with a 7.6% FCF yield. I did not subtract impairment or store closing costs as was the case in 2012.

While maintenance CapEx will be less than $18M even if it goes to zero the company makes $62.1M, giving the investor in CONN a 10.7% FCF yield. Definitely not a long but is it really a short? That clue may lie in an assessment of management.

Management

For all that I tried I couldn't find anything terrible on management. They have over-reserved provisions for bad debts in the past five years. Perhaps they use this to massage earnings, I don't know. It still shows conservative accounting for bad debts, which based on their use of re-aging is probably necessary.

If they are massaging earnings (again I don't know if they are) it would be under the direction of a new CEO, Theodore Wright. Theo was previously CFO at Sonic Automotive and left for personal reasons. It certainly wasn't the salary, as his base salary in 2000 was almost the same as his salary now($450,000). He owns over 175K shares and his only open market purchase was April 4, 2011 when the company was trading for $4.83 a share. With the benefit of hindsight it was a well timed purchase. 

His compensation is tied to adjusted operating profit and ignores costs for store closings and severance agreements. Grand total he made $2.9M in 2012. While it certainly would be nice to find something far more devious, I see nothing that's really out of line. It's rare to see a CEO actually really own a significant chunk of shares, especially at a small-cap so I can't really hold it against him. I don't think most executives understand how the market works anyways, but that's a topic for another time.

Why this stays in the "Too Hard" pile

I found myself going down rabbit hole after rabbit hole with this analysis. While the company is not cheap, there is nothing caustic here. I believe there is some flexibility being applied to their accounts receivable portfolio but it seems a stretch to say that anything is materially wrong. Since they no longer have the off balance sheet vehicle it is considerably harder to hide bad loans.

Also, just because CONN is offering credit to lower quality borrowers does not mean they are a ticking time bomb. Payday lenders are incredibly profitable and as long as there aren't interest rate restrictions I believe they will continue to make money. Whether or not 18% is enough to charge is the question. If you forced me to make a call I'd say the portfolio is break even at best over a 10 year cycle. That is just a gut level assumption.

The rise in promotional receivables and decreased down payment could lead to trouble down the road and let's be honest: the brick and mortar store is not profitable. Until I can find something more definitive I will stay on the sidelines. I am often wrong and could be here. No position.

AIG: A megacap behaving like a smallcap

I feel like this story has been presented ad nauseum. So prepare to throw up. Everyone knows about AIG and what an awful company it was. On the flip side, value investors are pouring over it. I think this would qualify as a Michael Burry "ick factor" stock. In fact, I think that everything is overblown and a margin of safety exists that may be greater than any other insurance company out there. More so than even my favorite, Aspen. 

There are four inputs to this: Insurance segment (Chartis/SunAmerica), management, the government, and everything else (all the auxiliary lines of business/wind-downs/IPOs). 


Chartis


Personal and Commercial insurance arm of AIG. They will not win an award for being the best insurance company but they are certainly one of the bigger ones. In Q1 2012 they wrote $8.8B worth of net premiums and had a pre-tax income of $910M. They underwrote to a loss, so investment income was $1.2B for the quarter.

They returned $1B to the parent in Q1 2012 and $1.5B in FY 2011. Management expects Chartis to return between $2.5-4.5B to the parent in 2012. This is off $124.9B of total invested assets as of Q1 2012.

While I'm not overly impressed with their loss ratios it's not a total train wreck considering all the catastrophes the past two years. I give Chartis a "C" for a grade (average for those conditioned to grade inflation).

SunAmerica

This is their retirement and annuity branch. In Q1 2012 it returned $1.6B to the AIG parent. In the Q4 2011 earnings call management expected SunAmerica to return about $2B to the parent. This is off $192.8B of total invested assets as of Q1 2012.

SunAmerica is big, well known, and probably here to stay. With a fair amount of ignorance I would apply a "C" to SunAmerica as well.

Insurance Segment

Combined Chartis and SunAmerica have total invested assets of $317.7B. Backing out debt and reserves and everything else one arrives at a BV of ~$83B. Management expects annual dividends from their operating companies to come in at $4-$6B. This would obviously correspond to an ROE in the mid-single digits. Just like every other insurance company, AIG has "aspirational goals of 10% ROE."

I believe that in today's low rate environment, coupled with the insurance industry surplus this segment is worth 0.9BV or $75B. Perhaps that will improve in the future but for now I'll stick with a bland 0.9 assessment.

Management

In Andrew Sorkin's "Too Big to Fail" I remember when Lehman Brothers, hoping for a liquidity injection, sent Buffett an annual report and other documents. Buffett read them and made a note of anything he didn't understand. Supposedly the report was littered with notes by the time WEB was done reading it. Confusing annual reports are not good.

Financial companies are opaque at best. Reading an annual report should clear and concise.
Reading AIG's most recent annual report was a breath of fresh air. Everything was spelled out. Does this eliminate long tail risk? Of course not. I can only hope that the team hired to clean up the mess created in the crisis will actually clean up the mess. They are experienced insurance executives and they seem committed to returning money back to shareholders. Up to $30B by the end of 2015.  

Everything Else

This will be a gross oversimplification but here goes.

ILFC: Looking at Whitney Tilson's presentation I think it's reasonable to believe that ILFC is worth book value. The S-1 for ILFC shows shareholder equity of $7.6B(pg 44).

Maiden Lane III: AIG received $5.56B on July 12th. They'll now get 33% of the remaining equity Conservative value = $5.56B

AIA Stake: Obviously this will fluctuate with the market. As of March 31, 2012 their remaining interest in AIA was ~$8.2B. With roughly 290M shares held this has declined in value to roughly $7.8B as of 7/16/2012.

I have ignored the other assets, such as United Guaranty, deferred taxes, etc. I think that "Everything Else" is worth $20-$21 billion. It could be worth more and perhaps in a serious market swoon/IPO bust it's worth significantly less. Again, my hope is that this is conservative.

The Government

The Treasury owns 61% of the shares outstanding. With roughly 1 billion shares and a break even price of $28.72 there's a worry that the stock price will be stuck at the break even price as the Treasury unloads shares on the market. Management knows this is the belief widely held.

"Conventional wisdom has been, as long as the treasury has a large position $29 is probably the ceiling for the stock.'' Robert Benmosche Q1 2012 Earnings Call

I have no idea what the end exit strategy for the government will be. Quite frankly, I don't care as long as they exit sometime soon, which is the stated intention.

So why do I believe AIG is a buy?

First, there is a skewed shareholder base. The Treasury owns 61% of the shares, Fairholme owns another 5%. Therefore, 66% of all shares are held by two groups. One group doesn't care that much about selling price. This would be an inefficiency. What fund manager would want to be caught owning AIG while the Treasury also owns shares?

Pretty much just Bruce Berkowitz and friends.

In small cap stocks large ownership stakes sometimes present inefficiencies. In this case we know that the Treasury wants to sell their stake. They are a forced seller. The exact details aren't known but as I've said earlier, we know they are exiting. Once the Treasury exits fund managers around the world can start buying. I believe this makes AIG act like a small cap that has a strange shareholder base.

If we believe that "Everything Else" is worth $21B and the core insurance business of AIG can make $3-$7B a year (3%-8% ROE) today's it's easy to see why Benmosche thinks that they can return $30B to shareholders in the next three years. As the Treasury exits shares should get bought back.

Will things change once the Treasury's stake drops below 50%? Yes I would imagine. Luckily we have the other banks that took TARP funds as a road map. They made it out OK and I think AIG will too.

The second reason I believe AIG is a buy comes from the margin of safety. The company is priced as if no other business segment ("Everything Else") has any value. Today's valuation gives the insurance segment alone a P/BV 0.67 ($58.7B MC $83B BV). I find it absurd that ILFC, ML III, United Guaranty and everything else has zero value. Even if it does you still get SunAmerica and Chartis at a discount.

Is that discount warranted? No, I think there is excellent clarity in conference calls, annual reports and public messages. Thanks to the Joe Cassano PR campaign, AIG is combed over by everyone. Luckily for us, the AIG of today is not the AIG of 2007. Once the government exits and non-core businesses are sold you're left with a large insurance company.

I think it's reasonable to value the insurance companies at $75B (explained earlier as 0.9BV) and "Everything Else" at $20B in value. AIG, in my eyes would be fairly valued at roughly $51 per share(1.875B shares 95B fair value).  

Conclusion

In 2007 AIG was spread out and had it's hand in numerous cookie jars. Many of these jars had mousetraps and as a result AIG suffered. The company is completely different now and poised to return $30B to shareholders in three years.

Buying AIG at a discount today lets investors participate in a clear path to accretive book value growth. As the Treasury exits I believe management will buy back shares. This will grow BV and enhance the upside while reducing the downside of government intervention. This will also lift the stigma associated with owning common shares. Instead of a government backed entity you will have an insurance company selling for about 1/2 book value that generates 3%-8% ROE on an annual basis.

While there are many risks to this investment I believe the clear path forward and the deep discount to conservative book value is a compelling opportunity. Long AIG