Often times, the market gets caught up in euphoric craze over the "next big thing," causing a group of companies to trade up in correlated fashion, irrespective of significant differences in their businesses. This momentum-fueled irrationality creates opportunities to make and lose a lot of money. Stocks related to 3D printing (e.g., SSYS, DDD) have picked up the mantle left by solar stocks, SSD stocks, internet stocks, etc., and have all traded into the stratosphere despite very significant individual differences. The ExOne Company (NASDAQ:XONE) typifies this madness and has no business being valued anywhere near the $50 at which it is currently trading. In fact, I can make a compelling argument that XONE is a $15 stock selling for $50.
The first thing you need to know about XONE is the company agreed to go public four months ago at $14-16 per share (check out the 424A filed with the SEC on January 29). Just to point out the obvious, insiders -- who presumably know more about the company and its growth/profitability prospects than external investors -- agreed to sell a piece of their business for more than 70% less than the value ascribed to it by the market four months later. To a common sense investor like me, this raises an enormous red flag.
Also, despite its recent IPO, XONE is not a new company. The company was founded by Larry Rhoades, who passed away in April 2007. While there is scant information available on the company's history, industry data suggest XONE has been manufacturing 3D printers since at least 2001. Further, a quick Google search of Mr. Rhoades turns up his Pittsburgh Post-Gazette obituary, which points out that, "The Ex One has grown to annual revenues of about $25 million and doubled employment to about 150 worldwide, including operations in Japan, Germany and the U.S."; the obituary was dated April 21, 2007. The revelation about ExOne having generated $25 million in 2006 sales is startling given that in 2012 (i.e., six years later) sales were less than $28.7 million. To be fair -- and in large part due to the capital infusion provided by the IPO (and subsequent investment the company can make in its sales force) -- management guided 2013 sales substantially higher ($48-52 million), but, again, we're talking about $50 million in sales and, if the gross margin and operating expense guidance are to be believed, $1.5 million in net income for a company with a $660 million market cap! At management's guidance, ExOne is trading at 440x 2013 EPS and 12x EV/Sales, and, no, those are not typos! With that being said, let's take a look at the business...
XONE's printers are enormously expensive, which significantly limits the size of the company's addressable market. XONE's printers range in cost from $100,000 to $1.5 million, and the average selling price has historically been in the neighborhood of $1 million each. That might sound well and good to 3D printing bulls who expect additive manufacturing to displace traditional manufacturing, but the vast majority XONE's business revolves around creating sand molds into which molten metal can be poured to create usable parts. Unfortunately for 3D printing bulls (and XONE), it takes about 16 hours for each print to harden, two full days for each print to fully cure, and each sand-casting can be used to create just one usable part! Two days for one part! Thus, this is not a technology that can be scaled into volume production/manufacturing. Some examples of XONE's printed sand castings are shown here.
(Notice the jagged edges and chipping. I'm presuming XONE chose some of their better examples for this trade show, but these look pretty rough to my eye.)
Some of the company's printers (M-Print, M-Flex) can also print metal (only steel, tungsten, and bronze at this point), but that process currently results in parts that are only 60-65% dense and in order to be solid must be hardened with bronze (also extremely time-consuming), which is a highly impractical outcome for auto and aerospace applications in which few manufacturers want all of their parts to be a steel-bronze or tungsten-bronze alloy. Some examples of XONE metal fabrications can be found here.
In my opinion, the most important thing to know about XONE is the company's business model is very, very different from those of Stratasys (NASDAQ:SSYS) and 3D Systems (NYSE:DDD) in that the company can generate very little profit from consumables; subsequently, the company's profitability will not scale along with its installed base. SSYS, for example, forces printer owners to purchase the plastic used to create 3D renderings from SSYS at enormous margins; SSYS sells customers a kilogram of basic ABS plastic (that might cost a couple dollars to produce) for $250-300 and encloses the ABS plastic within packaging (with an RFID chip) that must be authenticated by the printing machine, thus preventing users from using generic plastic. Therefore, SSYS might sell 250 printers over a given time period but sell consumables into its installed base of 5,000 (or whatever it is) printers at enormous margins (likely >90%) and benefit from the installed base growing over time. (This is the same strategy used by Gillette with its razors and Hewlett-Packard with its printers. It's frustrating as a consumer but enormously profitable for the company.) DDD uses the same model. The preponderance of XONE's printers, on the other hand, use sand to create 3D renderings. As management said in the company's Q1 call, "[I]t's tough to ship sand thousands and thousands of miles from one pit to another country." Thus, per the company's own expectations, XONE expects to generate just 10% of its revenue from the sale of materials and other consumables (at "modest" margins), which means the company's profitability must come from the sale of printers and the company's production service center (NASDAQ:PSC) business.
Realizing the tremendous lumpiness associated with selling million-dollar printers to businesses with large-scale prototyping/fabricating operations (an incredibly small target market), XONE began opening PSCs in 2009. Imagine PSCs as a Kinko's; you show up with a 3D CAD, and the PSC prints the object for you. ExOne opened its first PSC in 2009 and currently has five PSCs, each of which the company believes can generate $5-7 million in annual sales. Of course, despite having PSCs that have been open for multiple years, XONE has only one PSC that has generated even a million dollars of sales in a single quarter (something that happened only once), and PSC sales were DOWN sequentially in Q1. Regardless, the company claims that by opening up to 15 PSCs over the next few years, it will have created a "recurring revenue" stream that should reduce the lumpiness of its printer sale business. Let's think about the math on this for a second. Assuming XONE is able to open 15 PSCs, and they generate an average of $5mm in sales -- a figure none of its five current PSCs has come close to sniffing -- the company will generate $75mm in annual PSC-related sales. Assuming extremely generous 25% unit-level operating margins and a 35% tax rate, perhaps the PSCs, in a best-case scenario can contribute $12 million in net income. Remember, this is a $660 million company!
Given that I always try to buy/short stocks according to a financials-based analysis of intrinsic value, let's quickly model XONE's potential earnings power if everything goes according to plan. (Once again, just to reiterate my point, XONE likely won't approach these numbers for a very long time -- if ever -- but I always like to figure out how much a short can hurt me if the company executes according to plan.)
Here is what I am assuming:
- Management believes the company will sell ~30 printers this year; I'm assuming they push that number to 40 (which is currently maximum capacity), even though the company introduced a new S-Max late last year (at five times the speed of the former S-Print), which drove the big sell-through in Q4 and into 2013, and which I believe is a pace that will be tough to sustain. (Before 2012, the most printers the company had sold in a year is five, and the company had only sold five printers through the first three quarters of 2012.) The blended printer ASP should drop from around $1mm given that the company has introduced a couple lower-end printers.
- Management has publicly discussed opening 15 PSCs over time; the company currently has just five PSCs, with two planned to open this year; I very generously assume an average $5 million in annual sales per PSC, even though I believe XONE's best-performing PSC is only generating ~$3.5 million in annual sales. (Some PSCs have been around since 2009, so the low sales level isn't because they're new!).
- My model assumes $3 million in capitalized cost per PSC depreciated via straight-line over 10 years ($4.5 total for PSCs) and $5 million in depreciation from the non-PSC business (mostly the factory at which they manufacture printers), so $9.5mm in total D&A, which suggests 25.5% EBITDA margins (in-line with management's long-term guidance).
My intrinsic value analysis yields a price target of $15.25 per share, which, not surprisingly, is right in the middle of the $14-16 range at which management sold a stake in the company during the company's IPO four months ago. I would expect the stock to drift down toward that level as the IPO lock-up expires (increasing liquidity and decreasing short-squeeze volatility), sell-side coverage expands, and XONE's true earnings power becomes more evident to those buying the stock based on its association with more profitable business models with much larger addressable markets and higher margin recurring consumables revenue.
Disclosure: I am short XONE.
Additional disclosure: I have initiated a small short position on the stock based on my analysis, and I hope to add to my position as the cost of borrow decreases.