Atlantic Power – just a matter of time

When a company pays back debt, its enterprise value drops (or more specifically, the cash generation which leads to debt paydown is the cause of the enterprise value decrease because EV is market cap minus net debt, but I’ll insert this in before somebody comments on my illogical statement!). All things being equal, when material amounts of debt are paid back and the underlying entity is cash flow positive, it should eventually reflect an increase in market capitalization. This process sometimes takes a long time.

Atlantic Power today announced a revision to their credit agreement, which dropped the interest rate payable by another 25 basis points (to LIBOR plus 250bps), and if they can get their leverage ratio to less than 2.75:1 then it will go down another 25 basis points further. Atlantic Power had US$400 million in term loans outstanding in September 30, 2019 so this will result in somewhat less than a million a year in annualized interest rate savings going forward.

Reading credit agreement amendments (original April 13, 2016) might not be exciting, but sometimes a few nuggets of information here and there come out which are interesting. Section 2.15(d) of the agreement has the following debt paydown table:

The term loan component is extended to April 2025 which completely eliminates short term credit risk. The de-leveraging is mostly finished. The question is when the market will start to price in equity appreciation – after 2022 the company still has 12 power plants that are under power purchase agreements and generating considerable amount of cash flows.

When a CFO quits

Looking at Stuart Olsen (TSX: SOX). It’s been on my radar simply because I see their name in the fitness facility that I so happen to exercise in. They also (or likely one of their subcontractors) apparently botched up the concrete job in the City of Richmond’s new swimming facility, which needless to say, is not good for them.

Here is a relevant timeline:

September 9, 2019: SOX’s CFO resigns:

Stuart Olson Inc. (TSX: SOX, SOX.DB.A) (“Stuart Olson” or the “Company”) today announced the departure of Daryl Sands, Executive Vice President and Chief Financial Officer.

“I would like to thank Daryl for his over 14 years of service and wish him well in his future endeavors”, said David LeMay, Stuart Olson’s President and Chief Executive Officer.

The Company also announced the appointment of Dean R. Beacon as interim Chief Financial Officer.

…

Stuart Olson has commenced a search for a new Chief Financial Officer.

January 8, 2020: New CFO hired.

CALGARY, Jan. 8, 2020 /CNW/ – Stuart Olson Inc. (TSX: SOX) (“Stuart Olson” or the “Company”) today announced that it has completed its search for a new Chief Financial Officer.

Effective January 9, 2020, Bharat Mahajan will assume the role of Executive Vice President and Chief Financial Officer. Mr. Mahajan replaces Dean Beacon who has held the position of interim Chief Financial Officer since his appointment was announced on September 9, 2019.

Mr. Mahajan is a Chartered Professional Accountant and brings over 28 years of senior and executive level professional experience to the Company. He has a proven track record of generating value as a Chief Financial Officer. Most recently, Mr. Mahajan held the role of Chief Financial Officer at Daseke, Inc. (“Daseke”), the largest owner and leading consolidator of specialized transportation in North America. Daseke acquired Aveda Transportation and Energy Services Inc. (“Aveda”) in 2018. At the time of the acquisition, Mr. Mahajan had been the Chief Financial Officer of Aveda and was asked to take on the same role with Daseke on completion of the transaction.

“I am pleased to welcome Bharat, as our new Executive Vice President and Chief Financial Officer. His financial expertise and extensive accounting leadership experience in public companies will strengthen our abilities to execute on our growth and diversification strategies,” said David LeMay, Stuart Olson’s President and CEO.

January 24, 2020: New CFO quits, old interm CFO comes back.

CALGARY, Jan. 24, 2020 /CNW/ – Stuart Olson Inc. (TSX: SOX) (“Stuart Olson” or the “Company”) today announced that its Executive Vice President and Chief Financial Officer, Bharat Mahajan, has notified the Company of his decision to resign and that he has accepted a different position.

The Company also announced the appointment of Dean R. Beacon as Executive Vice President and Chief Financial Officer. Mr. Beacon previously held the position on an interim basis from September 9, 2019 to January 8, 2020.

The obvious question is – what the heck happened? Two weeks into the job and he’s leaving?

When a long-time CFO exits the company, it could be for a myriad of reasons, including the fact that he/she just has done their job for king and country and wants a graceful exit – nothing to do with the underlying performance of the company. However, when a new CFO is hired and they leave 12 business days after the fact, it leads one to speculate.

A few possibilities:

a) He came in, saw some huge ethical issues, and got out;
b) He came in, the CEO did not like his personality and/or work performance (isn’t an interview process supposed to pick up on this?) and effectively told him to leave;
c) He came in, realized he was over his head, and got out?

I can’t identify a scenario where this sequence of events portends a good outcome for a publicly traded company. They have to release their audited year-end financial results by the end of March.

Unfortunately, I can’t get a borrow on the stock.

The opposite of panic

Pretty much everybody wrote off this dinosaur of a company ages ago (especially during 2018 when new management tried to figure out how bad everything was before they started administering the foul-tasting medicine of structural changes and cost controls), and as a result, there is hardly anybody left that wants to send their supply into the market.

The risk of insolvency has been abated with a very aggressive debt paydown. It’s all about cash generation at this point – revenues are declining, but there will be a point where things will stabilize. When that will be is anybody’s guess, but it seems at present to be more likely than not at a cash generating level. Inevitably, you will have the computer traders and other passive vehicles jumping on board, and then it’s off to the races again. There will be some backing-and-filling of the stock price as you will have people waking up and hitting the sell button, but something to keep in mind is the extremely small float of 7.3 million shares outstanding will likely mean that the volatility will be upwards (demand-driven) rather than down. Today, for instance, somebody at around 10:30 (pacific) really wanted 2,000 shares of stock and it was enough to take it from 10.15 to 10.45 in a hurry.

What a chart of panic looks like

Company should be able to generate about $100 million in cash through operations in Q4, the thermal coal part of their market segment is steady, and coking coal markets look to be in-line.

CapEx will be elevated due to Leer South construction and thus share buybacks will be slowed down (to roughly $40-50 million/quarter, my estimate, compared to the $75 million they’ve been doing), but at that rate they still retire ~5% of the stock outstanding quarterly since each share they’re buying back at US$60 will result in roughly a 20-25% ROI given the estimated future cash returns they will earn.

The panic is generated through two sources: ESG-forced policy investors forced to dump stock in coal companies; and fears that low natural gas pricing will displace coal power generation. The first is a social construct that simply serve to fuel incumbency protection. The second is more relevant, but power generation is a very slow-moving industry where long lead times and up-front capital costs means that existing coal plants will continue to be economically productive for years to come. This does not factor in coking coal, which is half of the company’s revenues, and has nothing to do with the war on thermal coal currently being waged.

In the meantime, the underlying company continues to generate cash. The company itself is in a net cash position.

As long as the cash is being generated, one of two outcomes will occur. If the stock price is at panic levels like it is currently, share repurchases will be massively accretive to EPS and will elevate the stock price when the supply dump is finished. The other option is the company can stockpile cash and issue a special dividend – with a lower share count, this leads to a significant cash outlay per share.

I have no idea how long this supply dump will be, but it isn’t often when you see companies trading at 3x historical PEs being mass dumped – of course, companies are valued on the basis of future earnings, but there’s quite a large margin of error to work with given that ARCH is far, far, far away from being insolvent!

Canadian Newspaper Publishers – Torstar, Postmedia and others

Even though one would think with the S&P 500 and TSX being at heights that I would find the markets devoid of investment opportunity. While the Amazons and Facebooks of the investment world do appear to be expensive (and entirely propelled by deficit spending, federal reserve meddling, low interest rates and a good dose of TINA), the smallcap world, much to my surprise, has been full of plenty of research candidates, both old and new. I’ve been doing due diligence on various companies over the past couple months and have nibbled here and there. Nothing was as obvious as Yellow Media was in the 6’s in early 2019, but several items have received my interest and present reasonable risk/reward ratios.

I will write and disclose one of them simply because I have gotten my position (it is a very low percentage position in the portfolio, as in my minimum size to warrant opening anything) and I am not interested in accumulating more at lower prices. I will also caution that its liquidity is less than stellar.

I will piggyback on the post Tyler did with FP Newspapers (TSXV: FP) – well worth the read – he did a good job. Just be warned if you trade FP that you can move the stock price 20% with a few thousand dollars of volume!

It is well known that the traditional news publishing industry has been upended by the internet. Even Warren Buffet was caught flat-footed by this to some degree (he has made multiple comments on two-decade ago annual reports about the competitive position of single-community newspapers). However, I will make the claim that most of the damage in the industry is done. It is not completely over but the horizon is finally visible again.

In terms of publicly traded companies in Canada (on the TSX), we have the following:

* Torstar (TSX: TS.B) – notably owning the Toronto Star, Canada’s largest daily newspaper. Will write about them in more detail below.

* Postmedia (TSX: PNC.A / PNC.B) – National Post, and many prominent regional media, including the Vancouver Sun, Calgary Herald, Toronto Sun and Montreal Gazette. The stock, despite having over 90 million shares outstanding and a $120 million market cap, is very illiquid. They have an anchor around their neck in the form of nearly $250 million in debt and mandatory cash sweeps, contrasted with the trickle of operating cash flow they do generate.

* Glacier Media (TSX: GVC) – Owner of some 60+ local news media brands, although this is a subset of their other significant business offerings. Unlike Postmedia, the stock usually trades in a day, and the bid-ask spread is much more reasonable (pennies vs. dimes). Their Community Media category (which includes publications such as the Victoria Times Colonist) is approximately 60% of their revenues. They have been treading water financially, and have a very modest amount of debt on their balance sheet (about $20 million). They are, practically speaking, controlled by the entity that controls Madison Pacific (TSX: MPC.C).

* Québecor (TSX: QBC.A / QBC.B), which also owns the major French language publication (Le Journal de Montréal) and others in French language. Québecor is well diversified beyond its ownership of newspaper publications (its ownership of Videotron, for example) and really doesn’t fall into the “trading like trash” category of the three companies listed above.

I’m only going to look at Torstar in this post. This post has less quantitative rigour than my usual posts but I’ve done those evaluations off-line. Also, I’ve been less than comprehensive in writing the following analysis, but there have been plenty of other considerations taken into the scope of this.

Structure

The company has a dual class share structure. Its original founder and owner, Joseph Atkinson, who died in 1948, left behind the company to his successors who own the Class A voting shares.

The Class A shares (approximately 9.8 million outstanding) have voting rights, are not publicly traded and are owned primarily by a voting trust that joins together seven groups of shareholders. These seven groups (descendants of Atkinson) collectively hold approximately 99% of the Class A shares of Torstar and approximately 17% of the Class B non-voting shares of Torstar. They effectively control the nomination of the board. Class A shares can be converted into Class B shares. Class A shares cannot be sold in a take-over bid unless if the same offer is given to Class B holders.

The Class B shares (approximately 71.3 million outstanding) are freely traded, and notably Fairfax owns 28,876,337 shares or 40% of the class. Their last disclosed purchase was on November 9, 2017 when they purchased 9.4 million shares at CAD$1.25/share.

This dual class structure looks fairly typical, except for the following provision:

The holders of the Class B non-voting shares are generally not entitled to vote at any meeting of the shareholders of the Corporation; provided that, if at any time the Corporation has failed to pay the full quarterly preferential dividend on the Class B non-voting shares in each of eight consecutive quarters, then and until the Corporation has paid full quarterly preferential dividends (7.5 cents per annum) on the Class B non-voting shares for eight consecutive quarters, the holders of the Class B non-voting shares are entitled to vote at all meetings of the shareholders at which directors are to be elected on the basis of one vote for each Class B non-voting share held.

This creates a control incentive – the company must pay 15 cents per share in two year periods, otherwise Class B shares will get to vote for board directors (i.e. Prem Watsa will be able to obtain significant influence, if not control of the firm – his voting stake in this instance would be 36%).

The bulk of my shares were bought at 40 cents, which means if the Torstar board wishes to keep control, they have to pay a minimum 18.8% dividend at this cost.

On their Q3-2019 report, Torstar eliminated their dividends (it was 10 cents per year prior). You can see on the chart when this announcement occurred.

The board of directors stated they will review the dividend policy again in a year. I do not believe they will reinstate a dividend until there is obvious evidence of free cash flow, or Q3-2021, whichever comes first.

Financials

The following is a very broad summary. I’ve dived into the financial statements, but will pick on certain details. Print advertising is eroding at a very fast pace (roughly 23% from Q3-2018 to Q3-2019), and flyer delivery (which used to be a vector for advertisers to stuff more paper garbage into doors of homes that competed primarily with Canada Post’s unaddressed admail) is down 10%. Digital advertising and subscriptions are roughly level.

I do not expect there to be much recovery in print, but because revenues have already fallen so much, one can envision that the inflection point on the inverse “S” curve has been reached and that future revenue erosion will be slowing. Print advertising is still 30% of the total revenues. All of the advertising money has to go somewhere, and it is likely it will show up in the form of digital advertising (either on Torstar or thrown to social media), or “advertorials” and the like.

With revenue losses, cost containment becomes a much more challenging factor, and management has been trying to trim costs. For the most part their effort has not been sufficient in line with the drops in revenues, and there have been recurrences of “one-time” restructuring costs and so forth (the most recent going to be the shutdown of the StarMetro line of publications).

The other big component is the partially consolidated 56% ownership of Verticalscope, which is generating losses and the subsidiary has a $144 million debt which it is slowly chipping away at. Torstar has been less than transparent in terms of accounting for this investment, probably because it has performed so terribly. They were forced by the Ontario Securities Commission to change the manner in how they reported VerticalScope results. For the gory details, you can read Note 8 of the financial statements and the beginning of the MD&A document.

The consolidated balance sheet itself, however, is not in bad shape. The company has $52 million in cash, $9 million in restricted cash held generously in allocation toward an executive retirement liability, and no debt (the VerticalScope debt is non-recourse). There are significant liabilities on the books in the form of the employee pension plan, which is estimated to have a $127 million solvency deficit as of September 30, 2019.

This number might be a little scary, but it is not as if the pension plan is devoid of assets – at the end of 2018, there was $806 million in pension assets. There were $77.6 million in benefits paid that year.

The main point is that the company has some financial maneuvering room and time to work with. While the situation is clearly adverse, it is not at the point where it encumbers management’s ability to operate (unlike Postmedia, where the high interest rate debt is like an anchor around the neck of the whole company).

Intangibles

The large intangible aspect that makes Torstar alluring is name recognition – being a major media driver in itself, with obvious critical mass, provides value. There are analogs to this, the most relevant one being the decision that Jeff Bezos made when deciding to purchase the Washington Post in 2013.

The other intangible aspect is that the Toronto Star is obviously aligned with the federal partisan leanings of the existing government, which means it will continue to be a receptive economic vessel for the federal government. The journalism tax credit is one instance of this.

Competition

Direct competition – The Globe and Mail (owned by another historical family-owned media empire, Woodbridge Company) is the only other direct competitor in this space. They are effectively the two legacy national (English language – Québecor owns the French one) newspaper organizations.

Other than this, the other competition is through other domains – broadcasting news, internet, and independent publishers. The combination of these three has lead to a non-trivial erosion of the fundamental business, but this has been well explored elsewhere.

Sentiment

Bad. Really bad. Other than the negative dynamics of the newspaper industry, the suspension of the dividend was probably the last nail in the coffin for a lot of investors to finally bail out. I do suspect a lot of the trading since that announcement was fueled by tax loss selling. Q4-2019’s result is probably going to be quite poor with costs associated with shutting down StarMetro and there isn’t any good news at all other than the federal government subsidizing the business with digital media and journalism tax credits.

Valuation

The company’s Class B shares traded last Friday at 43 cents per share, which gives it a market capitalization of $35 million. In happier times (in 2011) the company traded as high as $15/share. It has spent most of 2019 under $1, and so far in 2020, it has been under 50 cents.

The question at the end of the day is whether the entity can sustainably generate cash. I believe the answer to this is yes. The question is how much the underlying business has to shrink in order for them to get to that point, and whether management can execute on structuring a leaner organization to doing so. That remains to be seen, but things right now are priced for a huge amount of pessimism relative to their market capitalization. The range of outcomes in my books are them going slowly to zero to being able to recover to a much higher market capitalization – when using a linear probability curve, the expected value is higher than 40 cents per share.

The big kicker is the ticking clock on dividend payments, which I think will give the board of directors incentive to tell management to get going. In the event they want to save themselves, they will give a 15 cent per share dividend by Q3-2021, which will mean I at least get paid to wait. If not, I’m sure Prem Watsa will have better ideas.

To repeat, I have a very small position in Torstar. If they pull off a miracle and get back to a $250 million market capitalization, it will be a welcome boost to the portfolio. If it’s clear that things are going from “really bad” today to “really really bad with no hope”, I’ll take a couple lumps on the head but it won’t cripple my portfolio by any extent.

If you do trade this, be warned that liquidity is not the greatest. It typically trades around $10,000 in volume a day, and the 2 cent spread you typically see in the stock represents a 5% price difference so selling at the bid and buying at the ask is very expensive if you are impatient!