Would you like to invest in a brick and mortar retail business trading at 8.0x TTM EBITDA and a 2.0% free cash flow yield that has posted SSS declines in five out of the last six years (the one positive year was a de minimis +0.1%), has grown its store count by 16% over the past five years yet seen EBITDA decline by over 27% over the same period, or worse, EBITDA per store decline by nearly 40%, has generated an abysmal sub-5% ROIC over the past three years, produces a low-single digit EBITDA margin despite having a supposed "monopoly" in its market, burned through $45m of cash last year, has seen its inventory per store swell by over 15% the past three years, and is on the verge of being completely disintermediated from its market? Probably not, which is why I continued to be baffled by the bull thesis for Barnes & Noble Education ("BNED or the "Company").
BNED is a contract operator of on-campus college bookstores across the United States. As of August 1, 2015, the Company operates 736 stores which reach approximately 25% of enrolled U.S. college students. BNED was spun out of Barnes & Noble, Inc. ("BKS") on August 2, 2015.
As far as I can tell, the bull thesis hangs its hat on two items:
1. BNED holds a monopoly over the campuses on which it operates, giving it a dominant competitive advantage
2. BNED has a lot of room to grow given nearly 4,750 college bookstores across the U.S.
After I dismantle this bull thesis, I'll tell you why BNED makes a compelling short.
On the first point, it's true, in terms of operating a college bookstore on campus, BNED has a monopoly. Once BNED partners with a college, it has the exclusive right to own and operate the only bookstore(s) serving that campus. What's not true, however, is the idea that BNED has a monopoly on selling textbooks to students. For the better part of the past decade, numerous competitors have encroached this market and taken share from college bookstores, including online retailers like Amazon, Chegg and BookRenter.com, student-run businesses that found an arbitrage in buying textbooks overseas and reselling them to students in the U.S. (a right that was upheld by the U.S. Supreme Court in March 2013), and even the textbook publishers themselves. Further, BNED's results are a far cry from what one would expect from a monopoly business - low single-digit EBITDA margins and sub-5% ROICs. So while BNED does has a monopoly on the brick and mortar campus bookstore, this has none of the benefits generally associated with a monopoly, and is therefore meaningless.
On the second point, again, it is true that there is room for growth in this market. But if I was a BNED shareholder, I'm not sure I'd want them to keep growing given their track record. For starters, BNED operated 624 campus bookstores in 2009 and in that year it generated $115m of EBITDA, or about $184k of EBITDA per store. At the end of fiscal 2015 (fiscal year end April 30), the Company operated 724 stores and generated $84m of EBITDA, or about $116k per store. So store count grew by 16%, EBITDA declined by 27% and EBITDA per store declined by 37%. That doesn't make me feel warm and fuzzy. On top of this, ROIC has been sub-5% for at least the periods for which balance sheet data is available; given a WACC surely greater than 5%, BNED's economic value add has been negative for the past three years. Ergo, each new investment in a store has destroyed value. Now, bulls will point out that BNED has been investing in a digital platform called Yuzu, and they blame Yuzu for the EBITDA decline. Again, not true. Stripping out all the Yuzu expenses, EBITDA per store has declined from $184k in 2009 to $150k in 2015, a near 20% drop. So Yuzu isn't the culprit.
Setting aside that store growth has been value destroying, and assuming as a shareholder you want the Company to continue to grow store count, the opportunity is greatly overstated. The problem is that investors are focusing on store count instead of enrollment. Operating the college bookstore at Arizona State University, with an enrollment of 75,000 students, is very different than operating the college bookstore at Alaska Bible College, with an enrollment of 38 students. Okay, that was just to have a little fun with the example, but you get the point, it's about enrollment, not store count. Bulls point to the approximate 4,750 college bookstores in the U.S. (per the National Association of College Stores), deduct the 736 operated by BNED, the 950 operated by Follett and the 200 that were operated by Nebraska Books (acquired by Follett in summer 2015), then take some haircut to adjust for the fact that there simply isn't a 100% TAM and arrive at a market opportunity that looks sizeable (maybe 1,400 stores). The problem is, this analysis is flawed. At present, BNED's 15% store count market share actually represents a 25% enrollment market share (as the Company likes to highlight). Follett's pro forma 25% store count market share is actually closer to a 50% enrollment market share. This leaves a residual 25% enrollment open market (vs. the 50%+ store count market the Company boasts); with some portion of this not suitable for or very unlikely to outsource, there is probably only another 10% - 15% of the market that remains for Follett and BNED to duke it out over. Because of this, not only is the growth opportunity much smaller than people expect, but also new store adds will be dilutive and demonstrate diminishing returns as incremental stores have lower economic metrics (lower ARPU, lower EBITDA per store, etc.).
On top of this, doing some simple math, it really doesn't even make sense for colleges to outsource their bookstores other than to offload the headache. Right now, the average college bookstore in BNED's system is doing $2.5m in annual sales, off of this, BNED pays the college a 6% royalty (a substitute for rent), bringing in $150k for the college. After deducting the royalty, BNED is making about $150k in EBITDA per store… so excluding the royalty, each store could do $300k in EBITDA, which means the colleges are losing half their potential income.
Given this understanding of the underlying fundamentals of BNED's business, it's clear to me the bulls have this one wrong.
Now for the fun part, the fact-based short thesis. I have done a ton of work in the college textbook space, including some investing in the distressed capital structure of Cengage Learning. I've been privy to conversations with industry experts, restructuring advisors and other investment professionals that have looked under the hood of this industry and have seen what's really going on. The simple fact is, college bookstores are being disintermediated and the future supply chain will be a direct-to-consumer model.
A little history. For anyone that's gone to college, you know textbooks can be expensive - I remember paying anywhere from $100.00 to $250.00 per textbook, needing 5 textbooks a semester, and selling them back to the bookstore at semester end for "scrap value." In total, a student today can spend up to $8,000 on textbooks over the course of a four-year degree. Now, $8k may seem like just a night out at Pink Elephant (I'm surely dating myself there) for the hedge fund crew reading this, but for today's average family, it is a tremendous amount of money, and in the face of perpetually rising tuitions, a real pain point. This pain point has compelled students and the market to find a solution.
It started about 12 years ago when some enterprising students discovered they could purchase the exact same textbooks overseas for 50% less, and resell them in the U.S., undercutting the college bookstores, while still turning a nice profit. (See: http://www.nytimes.com/2003/10/21/us/students-find-100-textbooks-cost-50-purchased-overseas.html). This didn't become a huge business, but it was painful enough for the college bookstore and college textbook publishers to wage a legal battle that went all the way to the U.S. Supreme Court, where they lost in March 2013. Then, as Amazon was getting its footing as an online retailer, it stepped into the ring and began selling textbooks, again, undercutting college bookstores because it didn't (doesn't) have to contend with retail overhead. But these solutions still weren't saving students enough money, so a new market emerged, the used and rental textbook market. Whereas the overseas arbitrage and online retailing markets were more a thorn in the side of the industry, the rental market had a seismic impact on the industry and threw two big players into bankruptcy: Nebraska Books in 2011 and Cengage Learning in 2013. Check out some of the disclosure from their first day filings:
Nebraska Book Company - June 27, 2011
"The Debtors have a long history of growth, but over the last several years, certain students buying habits shifted towards online textbook providers, including most recently online rental textbook providers"
"the Debtors have been unable to refinance due to (NYSE:A) the banks' unwillingness to finance Debtors given the Debtors' recent decline in financial performance and increased competition; (NYSE:B) potential lenders fearing the Debtors' financial performance would not rebound; and (NYSE:C) market perception that the growth of digital books will have a long-term negative impact on textbooks as it already has with general interest books."
"Several years ago textbook rentals were uncommon, either online or in a bookstore. However, the availability and popularity of textbook rentals has rapidly increased with online companies such as Chegg, Inc. and BookRenter.com leading that change."
"After further analyzing recent market trends, the Debtors determined that student demand for textbook rental was significantly increasing."
Cengage Learning - July 2, 2013
"Recent market trends have considerably altered the landscape of the educational materials business. One such trend is a consistent decline over the last decade in demand for new printed materials, which traditionally was the primary driver of profitability in the Company's industry. Consumers are increasingly opting to rent new materials, purchase electronic books, and, most significantly, purchase or rent used books. Approximately 40 percent of all consumer transactions in the learning materials market in 2012 were used book sales or rentals. Of course, each of these transactions is less profitable- or not at all profitable-to the publisher."
"Additionally, a recent Supreme Court decision permitting the practice of importing textbooks from foreign markets (where they are often discounted to reflect lower demand and relative standards of living) into the U.S. for resale is likely to increase the incentives for doing so and may put further pressure on domestic pricing."
Clearly, there was nothing arbitrary about the rise of the rental textbook market - students are compelled to save money and they are willing to go where the market leads them to do so, which leads to the next point. After losing the U.S. Supreme Court ruling and getting hammered from the growth in the used and rental textbook secondary markets, the big three college textbook publishers - Pearson (~33% market share), Cengage (~22%) and McGraw Hill (~18%) realized they needed to do something to regain control of their content and maintain the leverage in the value chain, and their solution is the eTextbook. Over the past several years, these three publishers have spent hundreds of millions of dollars designing and developing digital portals where students can purchase eTextbooks and related coursework. For confirmation, please visit Cengage Brain, Cengage MindTap, McGraw-Hill Connect and Pearson CourseConnect. These are all DTC portals that completely bypass the college campus bookstore. Also from Cengage's first day filings:
"Now, the educational publishing market has entered the early stages of a major transition from print business models to a greater focus on digital products, with digital market share growing as quickly as 20 percent annually over recent years. The move to digital began with the simple substitution of electronic versions of textbooks for the printed forms. Over time, digital products such as homework programs and interactive learning software have increasingly been paired and integrated with print materials. And in some cases, digital products are becoming a favored medium for learning materials in the classroom. As much as 15 percent of learning materials sold today are sold in digital format, including course materials, homework programs, and interactive and online learning platforms. All indications are that digital will continue to grow in importance in this market."
And from Cengage's 2015 Annual Report:
The market for educational products and services has traditionally been oriented towards the sale and consumption of print textbooks. While print textbooks still represent the majority of sales in the market today, digital solutions sold either as stand-alone or in connection with a print textbook are becoming increasingly important to educators and increasingly relevant to their decisions to adopt course materials for their classes. As a result, electronic delivery of pedagogical material through laptops or tablet devices is increasingly supplementing and replacing sales of print textbooks. We believe that these trends provide us with additional opportunities to deliver and monetize our learning solutions.
The growth in our digital business gives us access to a greater number of students and generates new sources of revenue from our existing adoption customers. In contrast to print publications, for which we receive revenue only the first time a product is sold, but none in connection with any resales, our digital products cannot be resold or transferred. We therefore realize revenue from every end user. Digital formats also free us from traditional publishing cycles, increasing our speed-to-market and affording us greater ability to tailor our offerings by course and even by specific faculty and student preferences.
We distribute our products primarily through "bricks and mortar" college bookstores and online book retailers which sell directly to students. In recent years, we have invested in proprietary online distribution channels. Through CengageBrain.com, our e-commerce site, we offer over 18,000 products in multiple formats for sale, including digital curriculum solutions, e-books, textbook rentals, print books, study aids, and supplemental materials, designed to enable students to choose solutions that best satisfy their individual needs.
The benefits of eTextbooks and accompanying digital learning solutions are incontrovertible and include, among many other items: (NYSE:I) a far cheaper cost to produce and bring to market than print textbooks; (ii) lower prices for students; (NASDAQ:III) better adoption rates for publishers supported by required coursework purchases and electronic portal submission of homework; (iv) untethering publishers from traditional publication cycles; (NYSE:V) increasing textbook and new edition speed-to-market; (vi) greater ability for publishers to tailor offerings by course, faculty and student preferences; (NYSEMKT:VII) lightening students' load by trading in a 30lb. backpack for a single tablet computer; (viii) the ability for students to take and store digital notes and stay better organized; (iv) the ability to hyperlink text within books for supplemental learning; (NYSE:X) the ability to create adaptive learning problem sets; and (xi) the ability to include videos and interactive modules within texts.
To solidify this point, consider the following example using the popular finance text, Principles of Corporate Finance, Brealey, Myers and Allen. Based on pricing taken from Chegg, the new print version retails for $229.49 and the eTextbook costs $94.00; the difference is $135.00 of savings to the student. Multiply this by five textbooks per semester, two semesters per year and four years and you get savings of $5,400. Of course, the savings aren't always this extreme, but even $3,000 of savings is huge. The value proposition is also compelling for publishers. The same text is sold wholesale by the publishers for about $160.00 (30% gross margin for the retailer), however, given a 40% adoption rate (Cengage estimate, reflects students sharing, buying used or renting), a publisher only captures about $64.00 of the potential sale, and at a 75% printed textbook gross margin, that equates to $48.00 of direct profit; by contrast, eTextbooks have a 90% adoption rate (homework submission required over a digital portal to which students only have access with a unique code that comes with purchasing the text) and an 85% gross margin, so even when selling the eTextbook for $94.00 ($66.00 below the wholesale print price), the publisher gets $72.00 of gross profit ($94.00 x 90% x 85%), a 50% increase per title over the print model. This is clearly a win-win for students and publishers, but an upper case "L" for college bookstores.
It is because of this eventuality that college bookstores are in secular decline and will soon be relegated to living off the sale of coffee mugs and sweatshirts.
But wait, there's more! The numbers of BNED tell a very different story than what's be circulated by Barron's and other bull thesis advocates:
- Same-store sales declined by 0.2%, 0.8%, 0.3%, 1.2% and 2.7% in fiscals years 2010, 2011, 2012, 2013 and 2014, respectively; and grew by a mere 0.1% in fiscal year 2015. Virtually all of the sales losses are attributable to textbook sales. To state the obvious, if the cyclical trajectory is up and your sales are down, you have a secular problem. For fiscal 2015, management is guiding to a modest 1% comp store growth, but as explained later, this is misleading because the Company is actually buying this growth. Even so, 1% comp growth is below core inflation and a pretty weak performance.
- As already discussed, even after adding back Yuzu expenses, BNED's EBITDA per store and total EBITDA has been declined since 2009, a five-year period that has been pretty good for most U.S. retailers.
- BNED generates low-to-mid single digit EBITDA margins; this is pretty terrible considering the near monopoly each bookstore should have on a college campus. In reality, these low EBITDA margins indicate an extremely weak position within the value chain.
- To ebb the loss of textbook sales, the Company has deployed a number of "creative" strategies, including pushing higher margin merchandise products (e.g. logo sweatshirts). As a result, SSS finally turned positive in fiscal 2015 and gross margin looks like its expanding. However, this is deceptive. In reality, the Company is buying merchandise sales using advertising and bloated inventory as its currency. While this has helped turn SSS positive and expanded gross margin, it has meaningfully increased SG&A expense, completely negating gross margin expansion. SG&A as a percent of sales, excluding spend on the Yuzu digital platform, was 16.0%, 16.8%, 17.7% and 18.9% in fiscal years 2012 through 2015, respectively. As a result, EBITDA margin has declined from 6.4% in fiscal year 2012 to 4.7% in fiscal year 2015, despite gross margin expansion.
- In operations management, there is a concept called inventory optimization, which seeks to identify the level of inventory that maximizes revenue and minimizes inventory obsolescence. Revenue is maximized by not missing a sale due to lack of available product; inventory obsolescence is minimized by ordering only the inventory that will sell before it goes stale. For retailers on a quarterly fashion schedule, putting this theory into practice is challenging and requires a careful balance because the cost of liquidating unsold, out-of-fashion items can be just as expensive as missing a sale. However, for a college bookstore that carries classic literature and emblematic coffee mugs, inventory obsolescence is a lesser risk. BNED has caught on to this fact and started to bloat its inventory in order to lessen the chance of missing sales. This helps push SSS growth, but at a tremendous cost to ROIC. In effect, a new, permanent amount of capital has been invested into the system that must stay there to maintain sales levels. Merchandise inventory (ex. rental textbook inventory) per store has grown from $357,000 at the end of fiscal 2013 to $410,000 at the end of fiscal 2015, or a 15% increase; over this same period, average weekly product sales per store has declined from $47,000 to $41,700, or an 11% decrease. BNED is buying revenue, which is a very slippery slope for a retailer. For Q1'16, inventory was up another 6%.
- In connection with expanding inventory, the Company continues to run a meaningful working capital cash flow deficit. Working capital was a $90m cumulative use of cash over the past four years. As the Company adds new stores, and fills them with bloated inventory, this working capital cash cash burn will persist.
- As a result of increased advertising spend, growing inventory levels, and declining SSS, BNED generates deplorable ROIC and ROE. Being generous and using the Company's market value of equity (~$700m) rather than book value of equity (~$750m) to bring down the denominator, the Company has generated an ROIC and ROE less than 5% over the past three years (as far back as stand-alone balance sheet data is available). Tying this into Stern Value Management's Economic Value Added ("EVA") analysis, with a WACC surely greater than 5%, BNED has DESTROYED economic/shareholder value for multiple years and will continue to do so.
- The Company has a significant deferred tax liability which has caused its effective cash tax rate to average 57.8% over the past five years. This means cash taxes are much higher than the GAAP taxes presented on the income statement. Higher cash taxes vs. what is presented on the income statement makes EPS and P/E multiples misleading, but more importantly, these cash taxes, tallying to over $30m the past five years, impinge cash generation and take value away from equity holders. It appears this cash outflow is currently running at $10m per year with the liability currently north of $70m.
Given all of these headwinds, value detractors and the Company's current valuation, I believe BNED is a very compelling short. The bulls will have their daily victories, but long-term, BNED will get crushed.
Disclosure: I am/we are short BNED.