Description

Saputo (SAP) is a global dairy processor (the 8th largest dairy company in the world). They make relatively-commoditized cheese, milk, butter and other dairy products; and yet, they’re trading at multiples that suggest they are a highly-profitable packaged goods company of the Nestle, Unilever, Danone, or Mondelez ilk.  Trading for 21x NTM earnings, well above its historical median and the average of its more appropriate ‘processor’ peers, SAP’s multiples also completely ignore the fact that its:

• U.S. business enjoyed short-term boosts to profitability last year, ones that are poised to mean-revert;

• Canadian business faces structural challenges;

• International business is struggling and will likely continue to do so.

Please note – all numbers in Canadian dollar, unless otherwise specified

INTRODUCTION

In 2014, with Class 3 milk and cheese block  prices at record highs, farmers globally expanded their cattle herds to take advantage of these record high prices, and the booming export markets prevailing at that time. But inevitably, as so often happens, farmers became too optimistic; and herd headcounts grew too far, too fast. And since it takes ~2 years for a baby calf to start producing milk, it was only recently that global meat, milk, and cheese markets begun to really feel the effects of this oversupply.

As cattle herds were growing over the last two years, a series of global events (Russia embargo, key European countries coming off a quota system, slowdown in China, strong USD hurting exports) compounded the problem for U.S. farmers in particular by drying up export markets and sagging demand.

Exhibit 1: Cheese Net Exports are plummeting

To make matters worse, U.S. farmers – buoyed by government “margin insurance program” subsidies – have done the exact opposite of what they should do in response to the mounting supply: they are producing more milk. With far more milk supply than the global market needs, and because milk does not store as well as cheddar which can be frozen, stocks of cheese have now mounted to unprecedented levels. The supply of total natural cheese stocks weighed in at a record high for the month of April (since the data was first recorded in 1917). A recent WSJ article highlighted how it would require every American to up their cheese intake by an extra three pounds this year to work off merely the excess supply of cheese out there.[1] This is probably a conservative estimate. With inventory of 1.19bln pounds of cheese currently in the U.S., and a U.S. population of ~324mm people, and if we assume cheese inventory turns every 30 days in a normal environment, that would require each American to consume 42 lbs. of cheese over next twelve months to work off these inventory levels vs. the ~30 lbs. that Americans actually consume per capita. We believe the prevailing cheese glut is likely to lead to continued pressure on cheese prices, and increased promotional activity at retail (and retailers are likely to try and pass off the cost of this to processors).

Exhibit 2: The “cheese glut”

Source: WSJ

Consequently, what a supply glut of cheese in the U.S. and elsewhere means for Saputo (SAP), is a brief period of margin boosting commodity deflation in their U.S. business, one which is now about to end, followed by a period where they have to pass through lower input costs to their customers, and a generally oversupplied environment at retail. The combination of these factors are likely to trigger a period of declining margins for SAP in their U.S. segment. This is doubly troubling since this segment has been propping their recent performance (more on why later).[2] The U.S. is Saputo’s largest geographic end market. To make matters worse, SAP also faces idiosyncratic headwinds in their Canadian, Argentinian, and Australian markets too.

Amidst this backdrop, and due to other idiosyncratic competitive headwinds facing SAP, ~C$1.98 in Fiscal 2018 (March 2018) earnings expectations is far too optimistic.

Further, at this point in the cycle, we believe the Street is extrapolating recent unsustainable boosts to profitability (lower input costs, possible earnings management and a strengthening USD - over 50% of SAP’s sales are in the U.S.). As these tailwinds reverse course over the coming quarters, consensus expectations will reset lower.

Given this, where SAP finds itself in the current dairy cycle, and the historical nature of the prevailing cheese oversupply, SAP should not be trading at 21x NTM earnings. This compares to a historical median over the last 10 years of 16x NTM P/E. We think something closer to a peak-cycle multiple is far more appropriate (min; max NTM P/E over last fifteen years is 12x; 24x). NTM EV/EBITDA is currently 13x vs. 10x historical median (and 7x; 14x min; max). SAP’s EV/NTM EBITDA trades rich by almost 5x turns relative to a subset of the more commoditized dairy processors (see in orange below), which trade at ~8x NTM EV/EBITDA currently.

Exhibit 3: Peer Multiples

The promise of both international acquisitions and industry consolidation are simply not enough to justify the multiples underpinning SAP’s current stock price. Further, we believe management is being vague about the drivers of their strong U.S. performance, and positioning commodity deflation-related, short-term boosts to profitability as the result of operational efficiencies and other supply chain initiatives, causing investors to be too optimistic. Also, in 2009 when similar dairy price deflation occurred in the U.S., SAP had more limited U.S. operations. For these reasons, we think investors are not properly accounting for the compression in margins that SAP is likely to experience in the U.S., once SAP has to start passing lower input prices to their customers.

Finally, in the face of the prevailing global “cheese glut”, after extensive channel work, we believe SAP is adversely-positioned at retail versus more branded counterparts: Sargento and Kraft, who have less-commoditized product portfolios than SAP (as evident in SAP’s relative EBITDA margins). Also, SAP’s positioning on the more commoditized end of the spectrum makes them even more vulnerable to retailers’ inevitable use of more heavy promotions, and to retailers’ ongoing effort to sell more private-label product.

INVESTMENT HIGHLIGHTS

Recent margin improvement in U.S. is unsustainable – SAP’s U.S. business benefitted from unsustainable boosts last year, including a favorable spread between cheese and milk prices, a strong USD, and the lag at which SAP passes through lower input prices in their Dairy Foods USA division.

SAP faces challenges in Canada, as that country’s managed supply system undergoes changes – With recent trade agreements opening up Canada’s managed supply regime to foreign competition, the artificially-high profits that dairy farmers and processors have enjoyed for decades in Canada at the expense of consumers is coming to an end. SAP’s and other processors’ imports of milk diafiltered products are further angering lawmakers in their home market, as it undermines the managed supply system.[3] Consequently, SAP has been and is looking abroad for inorganic growth opportunities. But so too is Agropur, their #2 competitor, who has been growing fast thanks to a consortium of institutional investors funding their M&A activity. The growing presence of Agropur is exacerbating SAP’s problems.

Saputo’s export opportunities and international expansion efforts have hit major snags – One of the major tenants of the bull thesis on SAP is their ability to consolidate the dairy industry through international M&A. But their track record in this department does not merit such optimism.

Earnings quality poor – Even in an environment of falling dairy prices, SAP’s Days sales inventory (or DSI) still increased dramatically y/y for much of the recent fiscal year. Their biggest increases y/y in DSI coincided with their biggest margin gains in their U.S. business. Given the large amount of fixed overhead in this business, we think the capitalization of fixed overhead contributed to their margin gains across 2015; and that it would be unwise to extrapolate such gains.  

VALUATION SUMMARY

We think SAP has ~57% downside from current levels. We see only 14% upside in our risk case, for a 4:1 risk-reward. We believe SAP’s topline will continue to struggle amidst low dairy prices, which we do not believe will improve over the next 18-months due to record inventory stocks, sagging export demand, and a lack of necessary production cuts. Further, we think SAP’s margins will come under pressure as temporary boosts to profit over the last year mean-revert, and as price competition at retail heats up due to an industry-wide inventory glut, and as the processors are forced to bear a disproportionate weight of this promotional activity.

COMPANY OVERVIEW

Saputo (SAP) is a Canadian-based cheese and dairy processor that was formed in 1954 by Italian immigrant, Guiseppe Saputo. SAP is the largest dairy processor in Canada (with Agropur now a formidable #2).[4] Their Canadian operations manufactures approximately 33% of all Canadian natural cheese. Saputo’s market share of total fluid milk and cream in Canada is approximately 36%. SAP is the second largest cheese producer in the U.S.[5]

Exhibit 4: Top Global Dairy Companies

In recent years, SAP has grown in the U.S., Australian, and Argentinian markets, largely through acquisitions.

SAP operates in three geographic segments: Canada, United States, and International (i.e. Australia and Argentina). On their most recent earnings call, they expressed interest in the Brazilian market as well.

SAP sells a complete range of: Italian cheeses, cheddar, specialty and fine cheeses. They also sell dairy products, such as: fluid milk, flavored milk, cream, yogurt, butter and butter blends and milk powder. Finally, they sell dairy ingredients, such as: Skim milk powder, Whole milk powder, Lactose, Sweet whey powder, Acid Whey Powder, Rennet Casein, De-proteinized whey powder and Whey protein concentrates. Saputo does not provide a volume or dollars sales breakdown by product type.

About 50% of sales go through retail channel, 40% through foodservice (mostly through broadline distributors like Sysco), and 10% to other food processors for manufacture in their products.

Canada skews more towards retail – with ~63% of sales in Canada going to retail. Canada is comprised simply of a Dairy Division (Canada). Brands in Canada include: Dairyland, Saputo, Armstrong and Milk2Go, and fine cheese brands: Alexis de Portneufand and DuVillage 1860.

In the U.S., retail is only 45% of sales. U.S. is comprised of Dairy Foods Division (USA) and Cheese Division (USA). Brands include: Frigo Cheese Heads and Treasure Cave.  

International is comprised of Dairy Division (Argentina), Dairy Division (Australia), and Dairy Ingredients Division.

Exhibit 5: SAP Corporate Structure

KEY INVESTMENT POINTS

Recent margin improvement in U.S. is unsustainable  

We believe the real reasons for SAP’s improved profitability in its U.S. segment over the last two years are unsustainable. Profitability gains have likely come not primarily from operational efficiencies or share gains, but rather from the unsustainable benefits of lower input costs and a stronger USD. These unsustainable boosts to SAP’s earnings will likely reverse course over the coming twelve months.

First, an overview of the “market factors” or drivers of SAP’s earnings, as they disclose them, is helpful in putting this part of our investment thesis in proper context:

• Market Factor #1: Average block market price per pound of cheese:

• And its effect on the absorption of fixed costs – This a simple, but perhaps underappreciated point, and is particularly relevant to a business like SAP that is asset intensive one (i.e. lots of manufacturing PP&E, which is about half of their tangible assets). Based on disclosures of a peer, we estimate about 20% of their COGS is allocated fixed overhead.[6]

• Absorption accounting is also relevant because SAP has arguably been overproducing product – i.e. as evident in their rising Days Sales Inventory for much of fiscal 2016. When average selling prices are lower in a given period versus in its comparable period, the portion of fixed overhead allocated to COGS becomes a larger percentage of SAP’s variable gross profit before any fixed overhead allocation. So all other things constant, declining cheese prices (like we saw in the calendar Dec. 2014 Q) means lower margins for SAP, because of absorption accounting, all other things constant.

• And its effect on the realization of inventories: Realization of inventories has to do with their inventory costing method (which is FIFO). As cheese prices fall over a given period by more than they fell in the same period one year prior, all other things constant, this will pressure SAP’s Cheese Division margins as they move the older, higher-cost inventory into COGS against each given sale, and they have to mark remaining inventory to fair value.

• Effect of the relationship between the average block market per pound of cheese and the cost of milk as raw material – Because milk is a raw material of SAP’s cheese production process in their Cheese division USA and of their Dairy Goods USA division (comprised mainly of their Morningstar business), SAP benefits from a widening spread between the Class 3 milk and cheese prices, and between the milk price and where they sell dairy good products. While these spreads can fluctuate in the near term – over time cheese and milk prices are almost perfectly correlated. This is because the Class 3 milk price is determined by a formula, which is largely driven by the price of cheese.[7] The Class III milk price formula is complex – but essentially determined by taking a free market valuation of things like cheese and butter (i.e. processed milk products), and then using these prices to determine the fair market value for milk. The price of cheese is the most important determinant of milk prices (as this is the biggest processed milk product there is).[8] The reason these two prices are so highly-correlated and changes in the spread between the two mean-revert is because if there was some huge discrepancy in the price of Class III cheese and milk (plus processing costs of turning milk into cheese), than every milk owner would arb this by converting their milk into cheese. This would cause an oversupply of cheese, bringing that price down (and the spread more in-line). So there is a market, self-correcting mechanism in the price between these two goods. So in periods where this Cheese-Milk spread turns positive (like now, early 2009, summer 2011 and 2012), SAP enjoys a near-term boost to margins, all other things constant, but one that will ultimately reverse, when the cheese-milk spread mean reverts. A similar dynamic is at play in their Morningstar business, which we discuss below.

Exhibit 7: Cheese-Milk spread at favorable levels; poised to mean-revert

Source: Bloomberg. Above statistics summarize data from March 2006-present.

• Market pricing impact related to sales of dairy ingredient – Another important product line for SAP is dairy ingredients (whey powder for example), as this is another option of what to process milk into. If the price of dairy ingredients rise relative to that for cheese, more dairy ingredients will be produced and sold by SAP within the limits of demand.

• Impact of the average butter market price related to dairy food product sales – There has been an increasing trend towards butter fat consumption in North America (i.e. as it has come to be viewed as more organic than margarine and other synthetic-type alternatives) that has caused butter consumption in the U.S. per capita to rise after decades of decline. Production capacity has not expanded with this rise in demand, causing an increase in butter prices. One interesting trend across 2015 was how North American butter prices did so well – particularly relative to international butter prices. But keep in mind, butter consumption per capita is about 1/6th the size of cheese consumption per capita in the U.S. Regardless, the price of butter relative to the price of milk helped SAP in 2015.

Exhibit 8: Performance of key dairy prices (and Class 3 Milk inputs) for SAP in early 2016 vs. 2014.  

During calendar 2014-15, milk protein and cheese prices declined dramatically, while the cheese-milk spread (particularly in calendar 2015) became more favorable.

With the above background out of the way, let’s turn now to the present environment. SAP actually discloses what change in y/y EBITDA they believe is related to the above “market factors”. But there is no way to verify if the amount they assign to these factors is actually true. And management leaves a lot of U.S. EBITDA growth “unexplained” – i.e. not a result of market factors, inventory write-downs, or FX translation benefits. So either management has created profit improvements that are long-lasting and sustainable – say due to the reduction of unused fixed overhead – or they are not being forthcoming with the true reasons behind their profitability gains in the U.S. business across the last 4 Q’s. We believe it is more the latter.

Exhibit 9: EBITDA y/y changes by segment

SAP’s fiscal year ends in March. So in Fiscal 2016, they grew EBITDA by $191mm in the USA segment. Woolwich acquisition was only a small contributor of this. They said that the above “market factors” were a $29mm headwind in their U.S. business, despite $82mm in FX-related tailwinds and inventory write-downs of $18mm. So this leaves $156mm of EBTIDA improvement unexplained.

Exhibit 10: Directional performance of each market factor by calendar year

First, to explore whether or not the EBITDA margin improvement in the U.S. business that began in earnest starting with Fiscal Q116 (June 2015) was due to more sustainable factors or not, we explore some of management’s commentary around the improvements.  

In August 2014, CEO Saputo highlighted the “closure of two plants in the U.S. specifically last quarter, which were executed within the first quarter of this fiscal year, our…New London plant in Wisconsin and our plant in Maryland. So again, all of that will contribute to increased efficiencies and better operations within the U.S. business.”

In Nov 2014, Saputo suggested that  following last year's capital expenditures in our Midwestern facility, and the closure of two U.S. plants, we've begun to benefit from lower operational costs.”

But EBITDA % did not improve y/y throughout Fiscal 2015. Improvement did not begin until Q116 (June 2015 Q) – 3 Q’s later, which is - not coincidentally we believe – also right when the cheese-milk spread started to really turn favorable.

In their commentary, the company does allude to operational efficiencies, better procurement of raw material, and other factors. From June 2015 Q (Q1 Fiscal 2016):

I can point to some of the things that happened in Q1, just better negotiation of our raw material input costs. We've capitalized on some of the folks on the Saputo Cheese side in the U.S. to help us negotiate new contracts on the Dairy Food side, which allow us to be more effective and more efficient at raw material costs. I think also in this quarter here, we saw some great benefit and some great improvements on a Dairy Food side in terms of raw material utilization, spoilage, waste, a lot of those little things that are second nature to us on the Cheese side now have been implemented on the Dairy Food side.

They are not talking about shuttering excess capacity – or cutting out workers. While some of the reduction in spoilage and waste could result in a permanent benefit, they really just seem to be saying that they had lower raw material costs, which they likely did (mainly because of lower milk prices, which they enjoyed a benefit from due to the lag with which this shows up in their selling prices). Ultimately these lower costs, though, will get passed through to customers, albeit with a lag.

SAP’s capex as a % of depreciation did come down a bit in the fiscal year ending in March 2015, which suggests they were perhaps getting leaner and more efficient through shuttering excess plants.

Exhibit 11: Capex has come down to what appears to be more of a maintenance level.

Capex guide for 2017 is way up. But there are expenses related to the implementation of a new ERP system included in both Fiscal 2016 and guidance for Fiscal 2017. Less these amounts, maintenance capex is ~$190mm, which is in line with deprecation.  

Also, net PP&E has not come down in a way to suggest SAP is achieving a leaner manufacturing footprint.

For the company overall, SG&A as a % of sales has actually increased slightly in 2016. So the efficiencies are not on the op ex side.

Beyond what appears to be a moderate amount of operating efficiencies, the real improvement in the USA business, we believe came through lower raw material costs. They have perhaps bucketed an unsustainable decline in raw material costs, one that they will ultimately have to pass-through, as a procurement efficiency.

We think the overwhelming driver of SAP’s strong performance in the U.S. over fiscal 2016 was due to falling milk prices, both through how falling milk prices impact:

1. The Dairy Foods (USA) division, i.e. Morningstar business they acquired from Dean Foods. This is because there is generally a lag at which they pass through lower input costs to retailers that purchase their extended shelf-life dairy products, and;

2. Cheese Division (USA), i.e. a more favorable spread in the price of milk of cheese.

Dairy Foods (USA) division, i.e. Morningstar business

We estimate that as much as $60mm of the “unexplained” EBITDA improvement in the U.S. could be from Morningstar temporary margin improvement, resulting from lower input prices, and the lag at which this gets passed through.

Morningstar sells mainly extended shelf-life dairy products (ESL).

We can try to understand this business better by looking at old Dean Foods (DF) filings during a similar period where dairy prices were plummeting (early 2009) to see what impact this had on the margins at that time, and whether it proved temporary or not. The filings from DF at the time do indeed show strong (and short-lived) profit growth over this time.

For example, in the Q ending March 2009, DF’s Whitewave-Morningstar business had a 320 bps improvement y/y in EBIT% (10.5% from 7.3%). In the earnings call Q&A:

But looking at the underlying businesses, was most of the drive in operating margin or EBIT growth from Morningstar because of the benefit from lower input costs?

Answer – Jack Callahan: It was a little bit more weighted towards Morningstar, but maybe 60-40 Morningstar versus WhiteWave. But both businesses had very strong flow-through on their businesses. Morningstar did benefit with some solid volume growth and an improving commodity input cycle that – versus what we had seen in 2008.

Milk prices were down dramatically in the Q, “The Class I milk price averaged $11.96 per hundredweight, 37% lower than the $19.12 average in the first quarter of 2008. Prices appear to have bottomed out in the near term, hitting a low of $9.43 per hundredweight in March.” And it takes a while for the pass through of these lower costs at retail to occur.

In the Q ending March 31, 2010, EBIT margins for Fresh Dairy Direct-Morning Star (note: they changed segment reporting) compressed to 5.1% from 9.1% “primarily due to increased pricing in response to higher commodity costs, primarily conventional raw milk, as those costs were partially passed through to customers.”

Morningstar was doing ~$1.6bln in sales, and $153mm in EBITDA, when it came over to SAP in Fiscal Q4 2013 (i.e. Q ending in March 2013). If we assume that similar dynamics as described above created 300 bps margin tailwind in just the Morningstar business using some estimate of today’s sales for that business (~$2bln), this would amount to a $60mm EBITDA benefit (one that is not sustainable, because SAP will ultimately have to pass through these lower input costs to their customers).

Below is the Class 3 milk price ($/cwt) from March 2014 – June 2016 and a period of comparable milk deflation (Mar 2008 to June 2009) to demonstrate the similarities between the two periods mentioned above.

Exhibit 12: Recent milk price declines reminiscent of 2008-09 declines

Cheese Division (USA)

We do not have the advantage of getting precise volumes by product line (cheese, dairy ingredients, and butter). But we do know that consumption per capita of cheese in U.S. is ~6x larger than that of butter. Whey protein consumption is even lower than butter. SAP’s volumes likely reflect a similar split.

Also, given we think Dairy Foods USA business does probably ~$2bln of sales, then that would mean the Cheese division account for not quite 2/3 of the sales for the U.S. segment.

So it is perhaps not surprising that there has been a strong correlation between SAP’s stock price and cheese prices (until very recently that is). Cheese is also a huge portion of their Canada segment.

Credit Suisse noted in a June 2014 report, “Correlation of stock price with cheese prices may also be supporting valuations. It's unclear if elevated cheese prices and such valuations can be sustained. Cheese block price in Q4/14 was $2.18 vs. $1.67 in Q4/13. Current cheese block of $2.03 compares to $1.78 in Q1/14.”

Well, those high cheese prices at the time were not sustained, as declining cheese exports, increased production levels, and consistent domestic demand ultimately weighed on the price of cheese.

Exhibit 13: Cheese block prices have sold off dramatically since 2014 highs

Fast forward to present day, and the historical correlation between SAP’s stock price and cheese prices appears to have broken down, as cheese prices fell but SAP’s stock oddly did not fall with them.

Exhibit 14: Correlation between cheese prices and SAP’s stock price has broken down

The data points circled in red are all recent periods – suggesting that SAP’s current stock price is elevated relative to cheese prices, when looking at the historical relationship between those two variables.

The correlation between cheese prices and the multiple the market is willing to place on SAP’s earnings has also been strongly correlated – until recently.

Exhibit 15: SAP EV/EBITDA multiple looks high relative to recent cheese prices

The timing of the above decoupling in SAP’s multiple and the price for cheese also coincided with a broader multiple expansion that occurred across a broad universe of “non-resource” Canadian stocks. As oil prices fell, and much of resource Canada became less attractive from an investment standpoint, significant amounts of institutional money poured out of resource stocks and into other sectors, like consumer staples. We believe SAP benefitted from this broad reallocation of capital – but that the inflation in their multiple that resulted will prove to be a temporary phenomenon.

Also, retail prices for cheese have been coming down since 2012. This is relevant to a processor like SAP, because as retailers’ margins get squeezed, they naturally put more pressure on their suppliers. SAP’s EBITDA margins have perhaps not coincidentally very closely mirror the trend in cheese retail prices (i.e. both peaked in 2012).

Exhibit 16: Avg. retail price under pressure as supply gains outpace consumption gains

The explanation for the decline in retail prices for cheese is similar to that for producer prices: domestic consumption of cheese per capita is plodding along at its historical trend rate, but export demand has declined – and production of cheese has ramped unabated, leading to too much inventory.

The question is why the relationship between SAP’s stock price and cheese prices has only recently broken down? We believe there are three reasons – and none of them sustainable or justifiable.

The first reason relates to the key profit driver for SAP working in their favor at present – the spread between milk and cheese prices. As the cost of milk (input) declines relative to cheese prices (output), this creates a temporary tailwind to margins as noted above. Notice how the temporary spikes in this spread in 2009 also correlated with an unsustainable boost to margins.

Exhibit 17: Cheese-milk spread and EBITDA margins closely linked over time.  

The benefit that results from this dynamic appears to exist up to about a ~9 month lag, before it mean reverts. We believe this time will be no different. The recent 2 Q’s of EBITDA margin improvement for SAP is likely largely a result of this temporary phenomenon. While this may persist for another Q or two more, consensus has EBITDA margins expanding out to 11.8% by March 2018 from 10.7% LTM. Against the backdrop of a likely declining spread between cheese and milk Class 3 prices, as the glut of milk and cheese continues to work through the supply chain, these expectations seem overly aggressive.

Further, lower cheese prices are likely to remain a margin headwind due to negative fixed cost absorption. This was a headwind in fiscal 2015, as Exhibit 10 above demonstrates.

Further, SAP will have to start passing lower input costs in its Dairy Foods business to customers. CEO Saputo said so much on the most recent earnings call:

So if I look at the U.S. model, I would say within a lag of maybe three weeks or four weeks, typically the milk price itself will balance out with the selling price of goods, so not a whole lot of impact when you look at the volatility other than the overhead absorption issues. In a higher market, we're able to absorb a little bit better the overhead. Where we have a little bit more of a lag would be in the international market, which not a direct correlation international price to milk price.

Further, in their annual report:

The decline [in cheese prices] in both fiscal years [2015-16] resulted in an unfavourable realization of inventories; however, the impact was more pronounced in fiscal 2015. The relationship between the average block market per pound of cheese and the cost of milk as raw material was favorable in comparison to fiscal 2015. However, the lower average block market negatively affected the absorption of fixed costs.

We believe the other ~$100mm of “unexplained” improvement very likely came from these cheese-milk spread dynamics.

The second reason we believe SAP’s stock price has decoupled from cheese prices is the strength of the USD, resulting in a boost in CAD earnings via the translation of SAP’s U.S. segment earnings back to their local currency (note U.S. is their biggest geographic region). In the absence of translation benefits, SAP’s sales y/y would have been down for the year – and down the most in Q1-2, when the margin improvements were most pronounced.

Exhibit 18: impact of strengthening USD vs. CAD dollar

Without the help from stronger USD, SAP’s EBITDA growth in their U.S. segment would have been 44% lower than it was.

The third reason we believe that SAP’s stock has decoupled from the price of cheese is due to the market’s belief that SAP will pursue lucrative M&A internationally with their under-levered balance sheet and low cost of debt; and thus insulating them somewhat from the challenges in the U.S. and Canada markets that we outline in detail above.

These temporary tailwinds combined are causing the market to extrapolate recent margin expansion. But we believe that the oversupply of domestic cheese stocks in the U.S., and the global price for cheese, will ultimately result in a declining cheese-milk spread and declining margins for SAP.

Because SAP did not own Morningstar through the 2009 dairy cycle, and because SAP is perhaps not being forthright about the impact of “market factors” on their results, the sell-side and consensus is largely missing the above. We have read some accounts that referred to the Morningstar deal as a “home run” – and validation that SAP is a “platform” ripe for acquiring other dairy processors and incorporating them on to their “lean” operating structure.

Is it possible that SAP has wrestled some efficiencies out of the integration? Surely. But we believe the overwhelming majority of the improvement in their USA business is due to unsustainable “market factors”, whether management will openly attribute the gains to that or not. Also, operational efficiencies, if present, should be showing up other places (lower SG&A/Sales, lower PP&E, higher ROIC, etc.). But this is not the case. ROIC, for example, has actually diminished since the Morningstar acquisition (granted also partly due to International acquisitions not going well – see section below).

Exhibit 19: ROIC has been under pressure since SAP’s spate of acquisitions, beginning with Morningstar  

As the U.S. business’s profitability mean reverts in the Q’s ahead, we believe consensus will come to appreciate the real reasons behind SAP’s recent profit improvements; and therefore will lower earnings expectations and the large multiple premium on their shares.

SAP faces challenges in Canada, as that country’s managed supply system undergoes changes

Canada is now SAP’s second largest market; and in its native country, SAP is facing a host of structural issues (ones that have forced SAP to aggressively look for growth opportunities abroad).

Canada’s dairy industry - unlike that in the U.S. - operates under a ‘managed supply’ system that establishes price-setting mechanisms, and limits domestic production and the amount of imports through imposing high tariffs.

The logic behind these controls is they allow governments to avoid subsidizing the industry directly, and instead to pass costs off to the consumer in the form of higher retail prices. Because the managed supply system creates higher retail prices, it is thereby necessary for the program to also include tariffs that block cheaper product from entering Canada.

This “managed supply” construct, however, is under attack in Canada in two ways: political pressure from international trade agreements and through the import of diafiltered milk.

First, many in Canada believe that the presence of the large supply management-driven tariffs greatly undermine Canada’s ability to forge new, constructive trade pacts with India, the European Union and the Trans-Pacific Partnership (TPP).  With regards to the TPP in particular, Canada’s initial refusal to negotiate around their supply management scheme within the TPP made Australia and NZ demand they be excluded from the TPP altogether. Canada ultimately capitulated, and agreed to make changes to their managed supply system, despite staunch opposition from the Canadian dairy lobby. Both the TPP signed in Feb. 2016 and the Canadian-European Union trade pact allow for dairies from Europe, New Zealand, the United States and other countries to now gain limited access to more of the Canadian market.[9]

According to Saputo, “The net effect on the dairy industry of all market access concessions on dairy products by all TPP member countries is expected to be slightly negative in Canada, neutral to slightly positive in the USA and slightly positive in Australia.”

We believe the potential negatives in Canada will likely overwhelm the positives elsewhere.

It will take years to phase in these changes, and this being after an extended ratification period. These changes are less about the immediate pain in the Canadian market they will cause, then it is about introducing more competition into the Canadian market over the longer-term, and how this will dictate where SAP’s future growth must come from: internationally through acquisition.  

While the supply management system has hurt consumers most (in the form of higher prices), and arguably all Canadians in the form of compromised trade negotiations, it has obviously benefitted not only dairy farmers, but also downstream processors like Saputo, as well as retailers, who collectively had shared in the protected profits created by the artificial competitive barriers the managed supply scheme creates within Canada.

The benefit to Saputo is limited to their domestic business, but at 35% of revenues, this is a significant portion of their top-line.

Lino Saputo in November 2015:

I think that dairy farmers are going to feel the pressure on this. If you look at the incremental access to the Canadian market, so you got about 3.5% from the TPP and perhaps another 3.5% coming from the CETA. So, you're looking about 7% of the total very solid access to Canada's market. Canada's consumption is not growing at a rate of 7%. So, somewhere along the line, there's going to be less milk produced in Canada. That's my perspective. So, it's going to be a hit to the dairy farmers… I'd like to say that if you look at Saputo's platforms, with our operations in Canada and our operations in the U.S. and Australia who are all three TPP members, we will find the right avenue to make this deal work for us… So, the first question that you had was related to the industry itself, I think it's going to be penalized. Saputo as a company I think will find a way to navigate through this

More specifically, the managed supply system in Canada increases Saputo’s cost of goods domestically. So the removal of this system, by lowering prices that dairy farmers can charge, will lower SAP’s COGS – so on the surface seems like it would be a good thing. But in Canada, SAP is able to pass their higher costs under the managed supply system off to the retailer, who passes to end consumer at retail. These costs can be passed on to the consumer because of the import quotas that protect these high domestic prices in the first place. And so, the managed supply system ultimately creates protected, outsized profits that are split up and down the supply chain (retailer, processor, and farmer) at the expense of the customer. One negative byproduct of the managed supply regime is that because of these same high prices, they are unable to competitively export their product.

We believe the removal of protected barriers to entry in their Canadian segment will therefore likely shrink the total profits shared between farmers and processors (to the benefit of consumers or retailers). The removal of the managed supply regime in Canada will be a net win for the end consumer and maybe retailer, and a net loss for the supply chain (farmer and processor). We believe farmers and processors are likely to bare the disproportionate brunt of this change, as the opening of Canadian borders will allow for retailers to engage in more direct sourcing from cheap overseas/producers. A recent report from Boston Consulting that looked at the potential impact of deregulation in Canada concluded by looking at international precedents that retailer margins benefitted the most from deregulation, whereas processors margins declined overall (in common product processing mainly, while holding up okay for branded and value-add products).[10]

Both because Saputo cannot get bigger in Canada in the first place due to their existing high market share/regulations, and because the changes to the managed supply system are introducing new levels of competition into the market, making it a less attractive one, Saputo must look internationally if they want to grow..

The second, and perhaps even bigger point to make about their Canadian segment, is the potential backlash that Agropur and Saputo might experience as a result of their recent increased imports of diafiltered milk, as a way to get around buying more expensive Canadian milk from Canadian farmers.[11]

Fights over this practice by dairy processors has even led to protests at Agropur plants.[12] According to The Globe and Mail, “Farmers also complain that dairies are abusing a duty deferral system to bring significant quantities of milk and milk ingredients into the country – a loophole that the federal government has now promised to close.” Emphasis mine.

It is impossible to say precisely what kind of profitability SAP has enjoyed from this practice of bringing in cheaper milk proteins (vs. more expensive Canadian milk), while still enjoying the high wholesale prices to retailers and their other customers due to the managed supply scheme. Some estimates suggest this practice has cost Canadian dairy farmers more than $230mm. Some speculate diafiltered milk imports have displaced 10% of Canadian milk consumption. So the impact of this game – which appears to be coming to an end – has likely had a material, nonrecurring contribution to SAP’s recent earnings.

Further, the “cheese glut” in the U.S. and the general dairy oversupply situation is likely to ignite these subversive imports by encouraging more permitted dairy and unpermitted diafiltered milk to leak their way over the Canadian border from U.S. exporters desperate to shed supply.

While the Canadian market will become more open to SAP’s U.S. and International businesses, SAP is far from the strongest in these markets. They are not guaranteed to gain from the opening of Canada in these segments, while they are almost guaranteed to lose from the opening up of Canada in the Canadian business.

One final point to make about Canada. Not only is SAP facing pressure from new trade agreements and scrutiny over its practice of importing diafiltered milk, but they are also facing intensified competition from Agropur, a co-op and Canada’s second biggest dairy company, that is now competing with SAP to try and aggressively consolidate the dairy industry in the U.S.[13] These two competing for deals, and the backing of Agropur by a group of Quebec institutional investors, make it unlikely SAP will win any deals without competition.[14]

Saputo’s export opportunities and international expansion efforts have hit major snags

A key tenant of the bull case on Saputo is that they will grow internationally, both through exports of dairy ingredients and through acquisition. According to RBC:

Earnings could be lower than currently projected if Saputo is unable to offset declines in cheese manufacturing profitability with incremental contribution from the production and sale of value-added dry whey and whey fractions.

Also, Saputo has a strong track record of realizing value-enhancing acquisitions and management continues to actively seek new opportunities, either in the US or abroad.

First, with regards to the opportunity of selling dairy byproduct ingredients, a steep drop in the prices for these products internationally has undermined this opportunity. From June 2015 (when the problems in that segment really begun):

I would also remind you that we do sell besides cheese, we sell ingredients as well into those markets, WPC and whey powder. And there's been anywhere from 25% to 45% drop in selling prices there and that has some impact on our Canadian business, that has some impact on our U.S. business and has impact as well on the international businesses that do sell value-added byproducts. So we are feeling the effects of the depressed markets, just about in all of our sectors.

Secondly, as of now the pursuit of growth internationally (name through entering the Australian and Argentinian market through acquisition) has not gone well.

Exhibit 20: International margins have been getting crushed

When SAP purchased Warrnambool Cheese in Feb 2014 (for $530mm), they did not disclose a level of synergies with this deal, but did hint at some:

I think that some of the synergies, to their own admission, is that they're making commodity products and they're price takers on the market. We are specialty oriented and perhaps we can make some products that would derive a higher value on the market. I think our sales force at Saputo has a very, very good understanding of the different markets, the growing markets, the specialty markets. And I think that there could be some synergies there just by sharing ideas and allowing that management team to flourish under some of our guidance and perhaps some of our fiscal support. Some of the products of categories are similar…So, I think there are going to be some synergies in what WCB does in terms of them adding value to the Saputo portfolio. And perhaps Saputo adding value to the go-to-market strategies that WCB can employ.

Saputo won Warrnambool after a protracted bidding war with Bega, and only after paying a substantial premium for the 88% of Warrnambool shares it purchased (note: Warranambool is a publicly-traded company).

SAP ultimately tendered for the shares at $9.40 AUD. The stock was trading for $4.25 AUD in Sep. 2013 before news of Saputo’s interest in October 2013. The stock currently trades at $8.64. According to Paul Jensz of PAC Partners, "Saputo has to find financial gains in Asia to justify a large purchase here as a beachhead acquisition. It's a stretch from a financial point of view."[15]  As Cannacord noted, “Saputo has indicated that it does not plan to rationalize or redeploy WCB’s fixed assets, and it would look to invest further to expand capacity.”  

Their third recent big acquisition was in May 2015 when a Warrnambool subsidiary purchased the cheese business of Lion-Dairy. They did not disclose any synergies for that deal. It was a $140mm AUD deal.

Also, ROE for Warrnambool plummeted to 1.8% in Fiscal 2016 (coincides with SAP’s) from 15.7% the year prior. So it is not evident that their past string of acquisitions (Morningstar, Warrnambool, EDC, Woolwich, Scotsburn) have resulted in any synergies as of now. It appears they overpaid for Warrnambool. And the gains in Morningstar are likely more due to unsustainable market factors, as discussed above, then they are the realization of synergies. Therefore, we are not sure future acquisitions will necessarily result in synergies either.

SAP’s outlook for ingredient markets does not suggest any improvement is on the horizon for this “growth” opportunity either:

The competitive market which existed in fiscal 2016 is anticipated to continue in fiscal 2017, and remains a Company challenge. Additionally, dairy ingredient markets have declined since the last half of fiscal 2015 and are expected to remain low through the first nine months of fiscal 2017. In order to mitigate downward margin pressures, stagnant growth and competitive market conditions, the Company will continue to focus on reviewing overall activities to improve its operational efficiency.

As such, the Company announced towards the end of fiscal 2016 the closure of three plants, being in Sydney (Nova Scotia), Princeville (Quebec) and Ottawa (Ontario). These closures are scheduled in June 2016, August 2016 and December 2017 respectively. The Division continues to leverage its operational flexibility to enhance profitability, in addition to maintaining cost control.

Saputo’s quality of earnings have been poor of late

Even after accounting for recent acquisitions, SAP’s Days sales inventory “DSI” has been on the rise. Given SAP accounts for inventory using FIFO, in an environment of rapidly declining prices, we would anticipate the higher cost, older inventory coming out into COGS relative to the lower cost, newer goods still in inventory to have a dampening impact on DSI.

Inventories/DSI did normalize in the most recent Q (ending March 2016) for SAP. But I highlight the elevated DSI over 3 of the last 4 Q’s to suggest this build-up in inventories could have helped margins in those periods in the U.S. segment. Perhaps not coincidentally, the increase in DSI y/y was most pronounced in the very Q that the U.S. business showed the most margin improvement y/y (Fiscal Q1 2016). In this period, DSI was up 24% after adjusting for possible M&A impacts, and EBITDA margin in the U.S. was up 410 bps above same Q in the previous year. SAP is about to anniversary this Q.

Exhibit 21: DSI – even after adjusting for M&A – was on the rise for much of fiscal 2016.

EBITDA margins expanded by 70 bps in Fiscal 2016. Consensus has EBITDA margins improving another 70 bps in fiscal 2017.

This extrapolation of margins is on top of what is already full margins relative to their peer set (orange section below):

Exhibit 22: SAP margins look very full relative to appropriate peer average.

Cash flow from ops (before changes in working capital) less capex from March 2011 to March 2016 has compounded at ~5% (from ~$520mm to $660mm). SAP currently trades at ~4% FCF yield.

VALUATION LONGFORM

In our upside case, we believe SAP shares are worth $16-$19 as of March 2017. We assume sales decline by 5% organically, and EBITDA margins compress ~80bps by the end of Fiscal 2018 to 9.9%. To arrive at our March 2017 target price, we apply median historical multiples (17x, 16x, and 10x NTM P/FCF, P/E, and EV/EBITDA) to our March 2018 estimates. We believe margin compression will result from the mean-reversion of temporary boosts to profitability in U.S. business, international struggles continuing, and Canadian structural challenges starting to gain steam.

We do not assume any further M&A in our model.

In our risk case, we believe SAP shares are worth ~$41-$44. In this case, we give SAP credit for ~4% topline growth, as depressed dairy prices start to rebound. We model margin improvement in-line with consensus, and apply a 25x, 21x, and 13x NTM P/FCF, P/E, and EV/EBITDA multiple (which is in-line with current multiples).

With these price targets, we see an attractive 4:1 risk-to-reward over a fairly short investment horizon.

RISKS

• SAP continues to grow margins in U.S. Segment through implementing continued operational improvement initiatives.

• SAP finds highly-accretive international acquisition targets in Brazil that they integrate on to their platform.

• Cheese inventories clear and cheese prices rise.

This report (this “Report”) on Saputo Inc. (the “Company”) has been prepared for informational purposes only. As of the date of this Report, we (collectively, the “Authors”) hold short positions tied to the securities of the Company described herein and stand to benefit from a decline in the price of the common stock of the Company. Following publication of this Report, and without further notice, the Authors may increase or reduce their short exposure to the Company’s securities or establish long positions based on changes in market price, market conditions, or the Authors’ opinions with respect to Company prospects. This Report is not designed to be applicable to the specific circumstances of any particular reader. All readers are responsible for conducting their own due diligence and making their own investment decisions with respect to the Company’s securities. Information contained herein was obtained from public sources believed to be accurate and reliable but is presented “as is,” without any warranty as to accuracy or completeness. The opinions expressed herein may change and the Authors undertake no obligation to update this Report. This Report contains certain forward-looking statements and projections which are inherently speculative and uncertain.

[1] http://www.wsj.com/articles/a-cheese-glut-is-overtaking-america-1463477403

[2] See Castor13’s well-written DF write-up. It sums up well a similar cycle further up the supply chain (for milk), and how it is impacting DF. We believe a similar dynamic is likely to play out for cheese processors as well. 

[3] https://www.dairyfarmers.ca/farmers-voice/dairy-products/what-is-diafiltered-milk

[4] http://www.dairyinfo.gc.ca/index_e.php?s1=cdi-ilc&s2=aag-ail

[5] http://www.cheesemarketnews.com/guestcolumn/2014/08aug14_01.html

[6] Fonterra 

[7] http://milkprice.blogspot.com/2009/04/is-milk-pricing-really-tied-to-cheese.html

[8] http://www.progressivedairy.com/news/industry-news/california-vs-fmmo-milk-pricing

[9] https://www.pehub.com/canada/2016/2/pe-backed-dairy-agropur-eyes-further-u-s-acquisitions-reuters/

[10] http://www.agropur.com/pdf/Analysis_of_%20impacts_of_supply_management_Canadian_dairy_inudstry-EN.pdf

[11] https://inuddernews.com/2016/06/

[12] http://www.theglobeandmail.com/report-on-business/rob-commentary/canadas-dairy-farmers-cant-stop-the-import-tide/article26854685/

[13] http://www.reuters.com/article/us-canada-dairy-idUSKCN0VJ230

[14] http://cdpq.com/en/news-media/press-releases/300-million-further-growth-agropur

[15] http://www.heraldsun.com.au/business/how-saputo-won-the-war-for-control-of-warrnambool-cheese-and-butter/story-fni0dcne-1226810019913

Catalyst

• SAP anniversaries the huge margin improvement/elevated inventory quarters from last fiscal year starting with the upcoming quarter.

• Forward expectations for cheese prices materially decline.