Saturday, 16 July 2016

United Guardian, Obscure Micro Cap, Great ROE, Great ROIC, Not Cheap Enough (Yet)

As I progress along my investing journey, I'm beginning to think in terms of not just attempting to find good and cheap stocks.  Both of these characteristics are transitory.  A company can be both good (based on return measures) and cheap (based on market valuation measures) at any point in time: It can continue to be good and cheap, and get cheaper, and cheaper, and cheaper (to the investor's chagrin).

Critical to developing a durable investment thesis is understanding what catalysts, changes, or under-appreciated events might eventually close the gap between the current state of cheap and an intelligent investor's conservative estimates of fair value (or range of fair values), in excess of current market value as measured by EV.

What's been missing from my analysis is an attempt to a) understand the significant current issues impacting valuation, b) determine possible catalysts which could sway investor focus away from current issues impacting valuation, and most importantly, c) develop an understanding of whether a company will be able to reinvest probable future cash flows and earn sustainable high returns on invested capital in excess of its cost of capital.

A company can appear to be super cheap (based on accepted conventional measures of valuation), but if it can't reinvest its future cash flows and earn sustainable high returns on invested capital in excess of cost of capital, it won't grow (in theory), and as a direct consequence, shouldn't appreciate in value.  The opposite is often the case.  I think this is why nose-bleed P/E stocks with extreme optimism surrounding consensus opinion on the way up tend to under-perform due to lofty expectations not being met.  Some call this P/E compression.  Expansion on the way up, and compression on the way down.  One of my favourite authors over at gurufocus had an article on Whole Foods Markets articulating exactly this, link here.

I find c) to be the most difficult concept to understand and exploit, because it requires peeling back the layers of current issues vs. probable catalysts, and forecasting normalized cash flows not currently appreciated as possible by the consensus while staying objective and realistic.  It's fairly easy to identify measures of cheap: it's fairly difficult to forecast how a company is going to overcome the "reasons" surrounding current cheapness (it has to be cheap for a reason), reinvest future normalized cash flows, and earn high returns on invested capital that very few other participants believe are possible.

The caveat to c) is developing a conservative appreciation of this potential (or range of potential) before investor consensus prices this in.  The best value investing opportunities seem to arise when the pendulum of pessimism prices in the demise (or diminution) of a going concern, and careful analysis reveals that this degree of pessimism is unwarranted.  The above approach is difficult to execute, because it requires sitting on your hands most of the time and being incredibly disciplined by way of not swinging at every pitch.  And my experience is, it;'s incredibly tough to sit on your hands because we, as investors, are led to believe that there are loads of temporarily cheap stocks at any one point in time.  This state of affairs is a direct consequence of bullish prevailing bias enveloping the financial universe.  Turn on the financial news, at any point in time there is no shortage of talking heads articulating a thesis as to why XYZ is currently cheap (at 40x ttm).  But, I've digressed enough. Onto the original subject of my blog:

United Guardian

I've posted about United Guardian in the past.  I owned a small amount earlier this year and got stopped out. For now, I have the company on watch.  It appealed to me when I first bought it because I thought that the company exhibited pretty substantial return metrics, and I based my purchase on these metrics alone.  But a cursory understanding of return metrics is often insufficient.

Over the last 10 years, the company has generated gross margins at or above 60% consistently, pretax operating margins of 44%, and net margins of over 30%, coupled with ROE in excess of 25%, and ROIC in excess of 25%.  I'd hazard a guess that consistency in operating metrics such as these would result in a a higher relative trading multiple, and indeed this had been the case up until recently (any by recently, I'm referring to prior to 2014).  I believed that high enough gross and operating margins over time were symptomatic of sustainable competitive advantage by way of pricing power.

As the company grew its top and bottom lines between 2006 & 2013, investors awarded the company a high multiple.  Between 2006 & 2013, sales and net earnings grew at a 3.4% & 11% CAGR respectively, even while the company experienced production shortages of its main pharmaceutical product between 2011 & 2013. The higher CAGR in net earnings relative to sales appeared to be a result of the company's ability to generate high returns on invested capital.  The company outsources the sales/marketing function attributable to its major product line, and so avoids the overhead ordinarily attributable to running a sales/marketing team.  At its peak in late 2013, the company was valued at 22x trailing earnings.

Since the end of 2013, something interesting has happened to ROE & ROIC, per below:

Between end of 2013 and now, both ROE & ROIC have dropped from 41% to 24% currently (ROE and ROIC are basically identical, the company has no interest expense so there's no difference between EBIT and EBT, so NOPAT  = NI)

Over the same period, revenue has dropped from $15.4M to $11.9M in the current ttm period, and NI has dropped from $5.9M to $3.67M .  While gross, operating and net margins have appeared to have held up, top and bottom lines are off by about 23% and 38% respectively.  There are a couple of possible explanations for the drop off in both revenue and earnings, more on this later. Suffice it to say, it has not been a fun ride being a shareholder between end of 2013 and now, see chart below:

While investors appeared to applaud the consistency of earnings growth between 2009 and 2010, the opposite has been true since the stock peaked in early 2014.

The company is tiny.  It's entire Enterprise Value is around $60M.  One insider, President and Chairman of the Board Ken Globus, owns 30.5% of the float.

The 2015 proxy statement is all of 17 pages, and the 2015 annual report all of 28 pages, link here.

The company is fairly easy to understand, from the 10K:

The company makes personal care and cosmetic ingredient products, under its Lubrajel products line, which accounted for 84% of revenues in fiscal 2015.  The company also makes pharmaceutical and healthcare products, the most significant of which is Renacidin Irrigation, which accounted for 9.6% of revenues in fiscal 2016.

The Lubrajel line consists of oil and water-based moisturizing/lubrication products, and from my understanding, these products are used as additives in the formulation of other personal care / healthcare products. The company relies on a marketing partner, Ashland Specialty Ingredients, to sell its lines of Lubrajel products.

The problems (as of late) facing the company appear to be as follows:
  • Reliance on the Lubrajel product line for a material (+80%) portion of its revenues
  • Reliance on a focused sales channel of this same product line into China (and exposure to any/all Chinese regulatory issues)
  • Intense competition by way of Chinese copies of its Lubrajel products in Asia and Europe
  • An inventory issue by virtue of its marketing partner, Ashland, accumulating excess inventory to last through Q2 2016.  As a result, the company previously warned that Q4 2015 and first half 2016 sales would decrease significantly vs. the first three Q's of 2015.  If the inventory issue is unresolved through first half 2016, excess unsold inventory plus competition could potentially result in pricing and margin pressure.  Certainly, this type of issue makes forecasting future sales difficult on the part of the individual small investor.

Based on the above, I can understand why the stock has come under pressure, and certainly, after having developed a better understanding of the issues potentially facing the company, I've relaxed my initial competitive advantage and pricing power assumptions.  A 22x peak 2014 P/E adjusted for a non-recurring spike in Ashland purchases of inventory quickly morphs into a 35x P/E after normalizing earnings to between $3M to $4M.

The question is, were net earnings of $5.9M an abberation due to Ashland accumulating excess inventory? And if so, what are go-forward normalized sales of the Lubrajel product line likely to be once all excess inventory is worked off?  With muddy visibility concerning growth prospects attributable to the Lubrajel product line, it makes sense that investors would tread cautiously until valuation reverts to a slow to no-growth state (i.e., P/E of 14x or less), suggesting further weakness in the price of the stock.

A Potential Silver Lining 

In 2015, the company obtained FDA approval to begin marketing Renacidin 30ml, an irrigating solution used to prevent and treat incrustations in urinary tract catheters, in a small, single dose plastic bottle.  (Ask me if I know what an incrustation is, good old google be praised, it's a formation of a crust).

All joking aside, prior to this FDA approval, Renacidin has been marketed and sold in a much larger 500ml dose format, and prior to 2010, accounted for almost 20% of sales.

The company experienced production issues with the 500ml format due to an issue with a 3rd party external manufacturer/supplier between 2010 and 2013, and according to the 2015 annual report, sales in 2015 were still "significantly lower than they were prior to those production curtailments, and the Company is continuing its efforts to try to recover some of the customers it lost as a result of those production curtailments and the consequent inventory shortage".  The theory now is that single doses will result in higher take rates and higher sales vs. the existing 500ml bottle.

Further comments from the 2015 annual report are reproduced below:

There are a couple of ways to look at this:

  • There could be potential pent up demand for the new single 30ml dose plastic bottles, which could translate into a jump in sales as promised post product introduction in April 2016.  The company may need to spend significantly on marketing the product though, and any adoption of the 30ml dose plastic bottle would naturally occur after the market absorbs the company selling off its remaining inventory of 500ml glass bottles (I'm not sure what the take rates or product usesage life of these 500ml bottles are).  In the company's Q1 letter to shareholders, the President noted that there should be no overlap between the two products, so I take this to mean that the 500ml bottles will have a short enough life cycle to have been used up once the 30ml bottles are ready for sale.
  • On the other hand, the company may not be able to fully replace the customers it lost previously unless the 30ml bottle really takes off.  And the problem I'm having here, I'm not sure how to even begin researching this!  In the company's Q1 letter, the President was pretty conservative in addressing the market potential, and made guarded statements along the lines of it being too soon to know whether the new single-dose form of Renacidin will have the hoped for impact on sales.

How do you estimate the market potential of the new single dose 30 ml bottle?

Really tough question, as there appears to be extremely limited information concerning pricing by product.  I was able to determine annual sales dollars by product for both Renacidin and Lubrajel going back to 2007 using a combination of extracting data from previous yearly 10K's and with reference to the following analysis of the company, courtesy of Value Investors Club, link here:

Conservatively, I'd guess that management would target at least a return to 2010 sales of $2.39M. Optimistically, though, I wonder if the opportunity isn't larger than $2.39M.  Critical to any analysis: will a 30ml single dose result in higher repeat useage?

I'm going to repost the author's analysis from Value Investors Club below.  Caveat, I have no idea where he got his 2.5-3.0x the price of the 500 ml bottle on a per ml basis assumption from.  I wonder if he spoke to the President or if the President addressed this in one of quarterly letters to shareholders:

So, conservatively, management will want to at least target a return to $2.39M in revenue, and optimistically, our author subscribes to the idea of $9-$10M from new 30ml bottles (in theory) in the medium term.

I've determined that there are 16.67 doses in a 500ml bottle, so does this mean that expected sales of the new 30ml doses are going to be 16.67x the old 500ml bottles in terms of unit sales (with no growth)?  In this case, I've estimated unit sales as follows, under two scenarios, 1) the company charges the same price per ml for the new 30ml dose plastic bottles it did for the 500ml bottles, and 2) the company charges 2-3 x the old per ml price for the 30ml plastic bottle (as per our author's estimates).  I've recast the historical results below (# bottles  = historical # of bottles x 16.67 in each case):

Objectively, with no per ml price increase per 30ml bottle, I've estimated a possible return to $2M - $2.4M in sales, and with a 2-3x per ml price increase per 30ml bottle, I've estimated possible sales of between $5-$6M.

I suspect that the company wouldn't have gone to all this trouble to get FDA approval if there was no opportunity in terms of pricing power in relation to its 30ml product.

If I take the average of the two ranges, I come up with close to $4M in potential sales if comparable unit sales in terms of 30ml doses return to pre 2011.

Tying it all together

The difficulty here is attempting to determine where Lubrajel line sales will stabilize.  Peak sales of $14M in Lubrajel might well have been a function of over purchases by Ashland, while pre 2011 Lubrajel sales averaged just under $10M.

If Renacidin 30ml reaches $4M in potential sales in the short term, the percentage of Lubrajel sales at $10M falls to 71%, and the percentage of Renacidin sales increases to 29%.  Under the assumption that the company is able to maintain trough gross margins of 56% and net margins of 26% (circa 2008), the company would potentially earn $3.6M, compared to expected net annualized earnings of $2M for fiscal 2016.  

At an average historical low P/E of 14x, I get a target value of $50.4M, less cash and securities of say $12M, I get an EV of $38M.  This works out to a target price of between $10 and $11 per share.

About that cash and final thoughts

A final comment about the $12M in cash, which works out to approximately 16% of total capitalization.  Surely, there are better uses for this cash than to keep it in the corporate coffers. Possible uses include: paying out a special dividend to shareholders, or an accretive acquisition to further diversify product mix.  

Another out of the box possibility? An eventual takeout by Ashland.  This would be a drop in the bucket for Ashland, who are already marketing the Lubrajel line.  At last check, Ashland had cash on hand of $1.3B.  United Guardian has an EV of $60M.  A takeout would be an afterthought, and if the stock got to $10 to $11 per share, I'd revisit the idea of buying the shares.

Sunday, 10 July 2016

Blog Update

I've cleaned up the blog a bit.  I've removed the links to the historical indexing study, the link to the shareholder account holdings, and the combined portfolios link.  If anyone is really interested in what I hold, they can just ask me. I'll be happy to oblige with an answer.  Suffice it to say, for the most part, I hold whatever I typically blog about if my analysis leads me to conclude that the subject security is worth owning.  The portfolios have some long standing holdings that I don't touch, but for the most part, I've sold most of the widely large cap names I held throughout 2015, and I've gradually replaced these holdings with buy and hold ideas that make sense to me.  I was pretty active during the market plunge in February and my regret today (hindsight being 20:20) is that I didn't hold onto some of the purchases I made in February longer.

My 52 week low process led me to buy some really great ideas at close to their 52 week lows, such as Boardwalk Equities and H&R, and CST Brands, but I took profits as they bounced.  This is a constant game of after the fact introspection, and there really is a lot of truth to the old adage of letting your winners run.  Holding onto just these three securities would likely have generated another 7% in return across the accounts (not including dividends).

I don't make concentrated allocation decisions in my portfolios, typically, I allocate between 1% and 5% of assets to an individual security or position.  I realize that after years of being a retail nobody, that I will most likely continue being a retail nobody and not have access to the same level of information and management interaction or candour that professionals have.  This could be a blessing and a curse.  I'm uncomfortable with more than 5%, but as the assets grow due to regular contributions, I have to look at a 5% allocation holistically from an overall portfolio perspective.  1% isn't going to move the needle much, if any.

The portfolios are currently structured as follows:

Total assets under management of $147,808

Cash = 42%
Debentures = 9%
Bonds = 32%
Common stocks = 10%
Preferreds = 7%

I'm always looking for new ideas using a combination of screens, candidates from my 52 week low list work, and catalyst oriented research.  There are a few really good authors I follow on seeking alpha who appear to be a source of objective research.  I also love reading Divestor and Prefblog for outside of the box analysis and thinking.

Final thought, if anyone is interested, I'll be uploading my evaluation models as a googlesheets link to the blog.  I just added my Whistler Blackcomb model (right hand side of the blog).  If anyone wants a copy of my evaluation model in excel, I'd be happy to provide it.

Friday, 8 July 2016

Whistler Blackcomb Holdings Inc. Only Game in Town Rhetoric is Getting Old

They are the only game in town. And at 10x 2015 EV/EBITDA, and 17x 2015 EV/EBIT for a consumer cylical (ttm), this story gets tiresome real fast (update, July 9th, realized I grabbed EV/EBIT at 17x, not EV/EBITDA - correction made).

Here's a typical example of the conventional rhetoric surrounding the thesis for owning the shares, courtesy of Baskin Wealth, link here:

The author has some interesting points, but this analysis hinges on one aspect of the investment thesis; pricing power.  Here are my thoughts after my own analysis:

  • Attendance has averaged 2,482 total visits between 2011 & 2016 (ttm).  Total attendance has been flat to lower every year since 2012, with a sizeable bounce back in the preceding ttm period due to the weakness in CAD/USD, which carried through to Q2 2016.

  • The business model seems to be predicated on maximizing and sustaining the formula, "price per visit" x "# of visitors".  If # of visitors falls, either due to overall economic sensitivity or "eventual" elasticity to price increases, can you really argue that a business has absolute pricing power?  Maybe, maybe not
  • A function of the business success has been the recent depreciation of CAD/USD. Therefore, there is a FX differential angle here.  US visitors have flocked to Canada rather than pay up to go to Vail. Similarly Canadian visitors have not visited Vail and have stay-cationed at Whistler. If price increases to the point of making US and/or Canadian visitors indifferent between Vail and Whistler, any FX differential angle will diminish, negating the pricing power argument.
  • A cursory look at the 2015 financials yields that the company earned $.53 per share, putting the P/E at 45x.  While I don't like looking at P/E in a vaccum (I prefer looking at P/E over a series of years), the P/E has been above 30 every year since 2012 (see morningstar graphic below). I interpret this to mean that investors have paid a premium for the the company being the only game in town since 2012. Translation, at 45x, there is little margin for error.  The company had better continue exceeding expectations.
  • I'm ignoring the forward P/E of 24x, why? Because it's misleading.  It captures ttm #'s without reflecting second 1/2 results which include summer months, and which historically, have shown losses. 
  • The author tries to make the case for a 7% free cash flow yield, 50% of which is paid out in dividends.  This argument is predicated on a sustainable dividend, more on this shortly.
  • The author closes with the comment, "we think this is a high quality company, with a long runway of growth, trading at a reasonable price".  My retail nobody question, since when is 17x EV/EBIT, 10x EV/EBITDA (July 10th correction) and an average P/E of 40x over the last 4 years "reasonable"?  

In the author's defense, I offer the following observation.  Looking beyond the historical average multiple of 40x, the author "may" be onto something.  The following is an excerpt from the Q2 2016 MD&A 

  • Per the above, the company is planning on a $345M additional investment over the next five to seven years in order to expand the park & build additional summer attractions.  Here's a link to the press release.
  • The expected impact and project phases can be found via the following link, courtesy of the company.

A few excerpts from the company regarding project renaissance, which is supposed to be transformational.

Key takeaways:

  • Renaissance will include building an all season adventure centre, independent of weather, an indoor sports complex, new outdoor non-skiing attractions, upgrades to existing facilities, and a new six-star luxury hotel
  • Renaissance will also include monetization of real estate by virtue of build out of a luxury townhouse development
So my retail nobody question of the day is twofold.  a) At 45x earnings, is any of this priced in yet? and b) How is the company going to fund a $345M development.

I can't definitively answer a), but I'll take a stab at b) as follows:

On the subject of free cash flows which the author alluded to above, here are the cash flows from the most recent MD&A for the year ended September 30, 2015:

And here is the cash flow statement:

Observations as follows:

  • Unless I'm mistaken, the company appeared to pay out almost 100% of free cash flow as dividends (not 50% per the author's insight).  Per above, CF ops was $70M.  They spent $30M on capx, leaving around $40M for dividends + debt service.  Dividends alone were $37M, which doesn't leave much room for debt service, unless dividends are cut, or capx is cut, or CF ops increases significantly (due to increases in pricing per the author's thesis I suppose).
  • With no change to dividend policy, the $.97 dividend with no growth (fair assumption as the dividend hasn't grown much if at all since inception), valued at 6%, 8%, or 10% perpetually is worth between $10 and $16 per share.  Ordinarily, I argue strongly against using DDM because DDM only values the discretionary portion of free cash flow as opposed to 100% of free cash flow, but as free cash flow seems to be almost equivalent to dividends paid, I feel DDM makes sense here.  Caveat, if a large %ge of capx is non-recurring in 2015 & 2014, then yes, free cash flow should be higher.
  • The company is planning on spending $345M to build-out the new features.  With a current credit facility of $300M, on which $127M + $75M was available as at March 31, 2016, they will either have to issue new equity or debt to fund the expansion.  See below:

  • Given that incremental positive cash flows from the build-out won't likely start rolling in until +5 to 7 years from project inception (and it's my understanding that all approvals have not yet been obtained), I'd hazard a guess that the company may need to revisit it's current dividend policy while stretching existing visitor cash flows in order to service additional debt (if additional debt is taken on to fund the expansion), or to pay additional dividends (if equity is issued), on the further presumption that visitor cash flows stay level throughout construction and construction does not interfere with visitor experiences
  • The valuation issue therefore becomes a function of attempting to PV the expected incremental cash flows attributable to the build out, and comparing this PV to the expected cost.  If this NPV + the existing operations' discounted FCF's exceed current EV, then the author is correct and 45x eps is not 45x eps (and valuation is reasonable).  
  • I'm not sure how to even start attempting to PV the expected incremental cash flows attributable to the build out, but I'm assuming there are some very smart managers running the company who have probably budgeted this all out.  They wouldn't take on a $345M project if it wasn't expected to yield a great IRR.  Here's my rudimentary attempt at determining a range of possible valuations attributable to Renaissance per share using EVA:

To start, I attempted to determine actual ROIC of existing operations, using normalized NOPAT / invested capital.  I note that historical ROIC has averaged around 5%.

In order to determine expected incremental cash flows from Renaissance, I used a range of expected perpetual incremental cash flows between 10% and 15%, less avg weighted avg cost of capital of 6.67%, discounted at a range of WACC's betweem 6% and 8%.  I then took the average of my perpetual incremental cash flows divided by # of shares o/s, in order to estimate a possible range of values reflected in today's price per share.  My estimates ranged from between $4 and $10.

As historical ROIC has averaged around 5%, I'm not comfortable blindly concluding that EVA/sh is around $10.  It just seems far fetched.  I'm leaning more towards the lowest estimate of $3.92 per share, and even here, I'm taking a leap of faith that ROIC will end up being double historical.

At $3.92 per share attributable to Renaissance, current pps of $24 attributes $21 to existing operations, and at $.53 per share, my P/E for existing ops drops to 39.6x earnings from 45x at $24. Not even close to reasonable.

(July 10th update, my discounted EVA analysis is overly simplistic.  It assume incremental cash flows start at year 1 and are discounted perpetually.  This is obviously not the case.)

Concluding Thoughts

To me, relying on the only game in town rhetoric as principle justification for owning the stock is dangerous, and at 40x earnings, buying the stock here is an exercise in ignoring risk.  Now, I am interested in the stock, but only at the right price, and only if an event unfolds which would serve to unsettle the existing shareholder base.  What type of event might achieve this result?  A dividend cut in an attempt to preserve cash flows during construction.  If such an event were to unfold, I'd be happy to buy shares as existing buy and holders rush for the exits, and only if I can buy the shares at less than 20x earnings.  Until then, I'm happy to stay on the sidelines and observe events unfold.  I may miss out on a consolidator event unfold (i.e., Vail or Intrawest does a deal with the company), but I'm not deviating from my principles and exposing myself to unnecessary valuation risk in order to ride on the coat-tails of every other shareholder who seems to share the same only game in town thesis as justification for sticking with the story.


Saturday, 2 July 2016

Empire Co. Ltd. Value Trap or Opportunity?

I haven't had time to write much over the last little while.  Work has been exceptionally busy, and there's been a dearth of beat up companies to analyze as the market hovers near all time, lifetime highs.

I've been looking closely at Empire Co. as it got demolished this past month.  Here's a monthly chart going back to 2000.  Notice that price has moved all the way down to what appears to be pretty significant trend-line support.  Now, strong caveat, just because I've arbitrarily drawn a trend-line doesn't mean it's going to hold (stockcharts doesn't have data prior to 2000, and I note that trading history has been longer than 20 years, which means any long term trend-line using earlier data points could be lower than the trend-line I've drawn).  What appears to be obvious support may prove to be the exact opposite.

In order to properly evaluate Empire, I have to understand why it's under-performed, what's wrong with the story, possible catalysts, and I have to attempt to do some sort of valuation (or a range of valuations).

What's wrong with the story (leading to under-performance):

In no particular order, here are my thoughts.  Empire is paying the price for a huge deal that seems to have flopped, and if it hasn't flopped, it has severely under-delivered vs. preliminary expectations.  

They bought Safeway in Western Canada, and in retrospect, not only did they seem to overpay, but Western Canada has been soft economically.  Their poor post-deal performance has only been compounded by the unfortunate circumstances of operating in a low margin, highly competitive business.  Empire seems to be losing ground where Metro and Loblaws are succeeding.  Empire was supposed to have fully integrated Safeway by now, but they haven’t, and they continue to spend on integration.  At the time, $5.8B was almost equivalent to Empire's capitalization, so what was supposed to be transformative has been anything but.  

I recall reading Bruce Greenwald's comments concerning generating incremental returns on capital invested. The best businesses are able to earn a higher incremental returns on each dollar of capital invested in excess of their cost of capital.  Any incremental dollars spent earning less than cost of capital result in destruction of value.

I certainly think this has been the case with the Safeway integration. For starters, there's little pricing power in grocery, so the deal economics (at the time) must have been evaluated on recognition of synergies as opposed to Empire being able to pass higher prices through to consumers.  I'm also hazarding a guess that no one within Empire at the time foresaw the oil crash in Western Canada (surprise, surprise), so part of the deal economics must also have been evaluated based on sustainable growth in Western Canada.  Ok, this all old news.

The major loss in the most recent year ended April 30, 2016 was mostly non-cash: they wrote down $2.8B of goodwill attributable to the Safeway deal, which works out to almost ½ of the purchase price for Safeway!  Put simply, they bought Safeway for $5.8B, and they wrote close to $3B off this year.

The drivers of the poor year over year results seem to be continued integration costs, unrealized or non-existent synergies, & continued softness in margins due to competition.


The pipe dream catalyst here in order for Empire to right the ship would be to divest of Safeway at a loss, stop the integration game and use the net proceeds to pay down debt, but given they are now almost two years post acquisition, and having spent significant sums on integration, this really is a pipe dream.  

Maybe, they could implement a program to convert more of their Sobeys or Safeway stores into discount stores, a la Metro or Loblaws?  After all, this seems to be a winning strategy for the competition, and Sobeys is already heavily discounting its goods in order to compete.

Another possible catalyst could be a full divestiture of Empire's 41% Crombie REIT stake, which at current market prices could bring in close to $800M, and with the proceeds pay down debt, or engage in additional sale leaseback transactions with Crombie REIT in order to monetize portfolio holdings and use the proceeds to pay down debt. 

Beyond these scenarios, I can't think of anything else that would really entice investors to buy the stock (or stop investors from selling the stock), but this doesn't mean that something completely unforeseen couldn't take place, i.e., privatization, merger/acquisition, sale of gas-bars or other assets, etc. (I'll offer that the dual share structure may limit the likelihood of a merger/acquisition taking place as voting control resides with the Sobey family).

This really boils down to a valuation analysis, i.e., is Empire cheap enough to merit consideration.  I find it fascinating that investors have no problem chasing Metro or Loblaws at 20x earnings for what appear to be similar businesses in terms of overall margins, overall pricing power, and competitive environment (at least on the surface), while shunning Empire the cheaper it gets.  See below, courtesy of morningstar:

What seems to separate Metro from Empire are differences in capital allocation and execution, i.e., Metro via Super C & Food Basics has executed well, while Empire via Safeway has not, and this has made all the difference in the multiple investors award each stock.  While Metro has been able to earn positive incremental returns on invested capital, Empire hasn't.


I approached valuation from two perspectives, 1) Sum of the Parts, and 2) Capitalized free cash flow.

Sum of the Parts

My rough sum of the parts analysis is as follows, going back to 2007:

My assumptions are as follows:

  • I valued Sobeys (food operations) separately and distinct from Safeway.  Safeway was added in fiscal 2014, so prior to 2014, I valued Sobeys at implied valuation on privatization in 2007 of $2.8B (Empire bought out the remaining publicly traded stake in Sobeys at $1B and owned just under 2/3rd's of Sobeys prior, so $1B / .36 = $2.8b)
  • I assumed Sobeys grew at the rate of net earnings from continuing operations each year between 2007 and 2014.  This would put Sobeys alone at a theoretical 5.5B by 2014 just prior to the Safeway deal
  • I determined 42% of Crombie's mkt cap each year between 2007 and 2016
  • Finally, I used the Safeway acquisition price starting in 2014 and reduced it in 2016 for the impairment charge of $2.8B in 2016
  • I determined estimated value per share attributable to Sobeys, CRR, Safeway, and deducted total debt per share to arrive at an estimated sum of the parts valuation.  I ignored other real estate investments (Genstar) and the previously disposed of Empire theatres in my analysis.
  • I then compared avg price per share in the market vs. my sum of the parts analysis in order to assess whether Empire on a sum of the parts basis has traded at a theoretical premium or discount to price per share over time.


  • It's interesting to compare my theoretical SOTP vs. avg stock price.  The two metrics have tracked pretty closely since 2007, per below.  As the business grew and generated positive returns on invested capital, the spread between theoretical sum of the parts vs. avg stock price remained positive.
  • As integration and negative return on capital issues with Safeway have become problematic, the spread between the two measures has inverted.  By my measures, PPS at just under $20 is lower than theoretical SOTP by close to $6 (this doesn't mean that this spread can't get wider)
  • The discount of theoretical SOTP to PPS at .71 is the lowest discount over the last 10 years.

What if I just ascribe a value of $0 to Safeway?

Here, I get a theoretical SOTP of $16 per share, ex-Safeway.  We're only about $3 or 17% away...

Free Cash Flow to the Firm

In this case, I used free cash flow as reported in each annual report going back to 2007, and I normalized the 2014 free cash flow downward by $640M in order to remove the non-recurring sale of manufacturing facilities.  I then added back after tax interest expense each year to get a better approximation of free cash flow to the firm, and I discounted the resulting free cash flows by 8% perpetually each year (no growth), and compared the results to EV.

For my purposes, I approached free two ways, 1) using as reported free cash flows in 2016, and 2) taking the average of reported free cash flows between 2012 and 2015.  Arguably, 2014, 2015 & 2016 include non-recurring Safeway integration and system related costs, so it's possible that normalized, sustainable free cash flows once the Safeway integration is done (or undone), should be higher than reported.

My results are as follows:

In this case, I get capitalized free cash flows of between $6.2B and $8.8B, vs. current EV of say, $7.3B.  If normalized sustainable free cash flows are closer to $630M, I'd argue that Empire looks about 17% undervalued here, but until the Safeway issues are fully resolved, I'd hazard a guess that the stock will continue to languish.

Concluding Thoughts

I believe the shares will continue to under-perform until there's better visibility regarding Safeway. While compelling on the basis of sum of the parts or capitalized free cash flows, these two theoretical measures alone don't guarantee that the stock won't continue to get cheaper.  I do note that at $16, it seems that you're getting Safeway at $0 on a SOTP basis.