Years ago the KGB had a research unit that studied fairy tales. It wanted to understand why some stories are so powerful and believable, and use this knowledge to improve its propaganda tools.
I have been following Amazon.com (AMZN) for years, and it does feel like a great fairy tale. Amazon's story has a very likeable main character who uses his wits to defend his kingdom (the company) from the beasts both large (Wal-Mart (WMT), Google (GOOG), Apple (AAPL)) and small (the critics who demand profits), and who would eventually lead it to the happily ever after (a respectable net margin).
Amazon.com is a fantastic company: dedicated, innovative, even daring. I am grateful that its website shows me as an author of 98 books that I did not write (it's a long story), and I am a very happy customer. I do, however, believe that the moment is approaching when investors may no longer be willing to support the current valuation of the company.
Shares in Amazon.com are trading at P/E ratio of 360, or 108 using the next year's expected profits, and the earnings multiple was at these levels for most of 16 years since the company went public. Very few companies in history managed to keep their valuations so high for so long. Investors believe that at some point in the future the company will be able to translate its rapidly growing sales into substantial profits. I have a problem with this logic: the future has already come, and more than once.
The Scale Hypothesis
The original idea was that as the company grew, it would be able to realize better margins through the reduction in customer acquisition costs and the improvement in operating efficiency. The following statement was a typical refrain in most of the reports from the early 2000s:
This CQ4 , with its distribution centers at more efficient capacity, with significantly improved vendor management, merchandising, inventory, and shipping practices in place, we believe Amazon should be able to meet our $1B revenue estimate (up 48% Y/Y) with significantly better operating results...
Year after year the company demonstrated phenomenal growth. Last year Amazon brought more revenues in a week than it did in the holiday quarter of 2000. Yet, the business was still unprofitable.
I believe that if the company could not translate a 27 times growth over the past 13 years into substantial profits, it never will. The opponents claim that the company made a conscious choice to pursue growth instead of profitability, and to reinvest profits from its older businesses into the new opportunities. I don't dispute that. I have a problem with the second part of their premise - that the company is able to crank up the profits whenever it finally decides to do so.
First, most of the benefits of scale have been realized by now. Growing from 1 warehouse to 5 cuts delivery times dramatically. Adding 7 new facilities (announced plans) to the existing 49 distribution centers in the U.S. will not result in the same rate of improvement. The same logic applies to all the other areas of customer experience and internal operations.
In fact, Amazon may have reached the point where the economy of scale turns negative, i.e. further growth leads to more expenses, not less. For example, the company has many more markets to support - both in geographical and in product terms, more business units to manage, and more platforms to develop and maintain. It used to sell books in the U.S. Now it sells everything from groceries to paintings, operates on several continents, develops its own hardware and software, streams movies, offers software services, etc. It is a valid growth strategy, but this fragmentation makes it very hard to substantially improve operational efficiencies from the current levels.
Look, for example, at key operational metrics - they are declining. While revenues increased by 150% between 2009 and 2012, the net margin turned negative.
The inventory turnover went from 12 times per year to 9 times, and cash from operations - a key measure of business efficiency - declined from 13.4% of revenues to 6.8%. These numbers are not fluctuations, they represent a clear trend:
Second, if the competition survived Amazon growing by 27 times, it is not going anywhere. Amazon.com will never reach the Never-never land where it stands unopposed and has the power to charge its customers whatever it wants. Some competitors did fail or faded away (Borders, B&N (BKS)), but the rest are still standing (Wal-Mart, eBay (EBAY)) and even growing. In many areas the company faces tougher competition today than ever before (e.g. media, tablets). There are hundreds of online retailers, like BHPhotoVideo.com or Tennis-Warehouse.com, that carved out their own niches and seem to be doing fine.
Any attempt by Amazon.com to raise prices or scale back customer service (including free movies, free delivery, etc.), will give a second wind to all the competitors that collectively represent a formidable threat. Selling merchandise online is a low-margin business, and it will remain so in the foreseeable future.
The bottom-line: a further growth in revenues, by itself, is not a game-changer. There is only one way to increase its profitability: the company must increase prices or scale back its customer service. This will lead to a slowdown in growth or even a decline in revenues. A growth in earnings will be offset by a rapid decline in the valuation multiples. The net result is likely to be a drop in the share price.
This realization has not set in yet. Even analysts, who watched the company underperform relative to their earnings expectations year after year, still tout revenue growth as the key catalyst for the stock.
The Razor Blades Hypothesis
As Amazon's sales zoomed up but failed to produce a respectable net margin, the company started selling its own hardware (e-readers, then tablets). This new twist in the Amazon's story captured the imagination of investors and analysts. They reasoned that Amazon is selling its devices at a loss, but these losses will be more than offset by sales of digital content to the owners of these devices.
This assumption has already proved to be wrong. Most buyers of the original Kindles have already upgraded to the newer versions, so any loss on sales of those units was never recovered. More importantly, by now consumers own tens of millions of Amazon devices, and the company already offers a huge selection of content, but the profits never materialized.
The razor blades business model did not work. There are several reasons for this, and they are not going away:
- Major competitors - Apple, Google, and now Microsoft (MSFT) - offer more than hardware. They offer the entire ecosystems that include their own operating systems, popular and familiar services (email, calendar, productivity apps), as well as complimentary devices (from smartphones, PCs and game consoles to AppleTV and Chromecast). It is very hard for Amazon.com to compete on all fronts. It has to offer its hardware at lower price points to keep the volumes high.
- Consumers turn increasingly to the all-you-can-eat services like Netflix (NFLX), Hulu, Spotify, Rdio, Google Play All Access, Xbox Music, iTunesRadio, Sirius XM, etc. For example, this year will be the first time when digital downloads of music decline, while the revenue of music streaming services is up 59%. Many of these services offer free ad-supported options. Many of them operate at a loss (Pandora (P)) or with a tiny margin (Netflix' net margin was 0.5% last year).
- Delivering quality user experience for digital content is expensive, after the costs of technology and customer service are factored in.
As a side note, Amazon's attempt at producing original shows is a side-show (pun intended). This is a very different - and risky - business that does not deserve the multiples that Amazon.com currently enjoys. This is also not an easily scalable business: it's much harder to produce 20 hit shows than 2. Amazon is forced to follow the lead here, but it is unlikely to generate profits big enough to make a difference for Amazon, if any.
The bottom-line here is that Amazon was not able to turn any profits from digital content so far, and the road forward is tougher, not easier. Most likely, the hardware/digital content combo is losing money now, and there is not much Amazon can do to change that. And it cannot exit this business either.
The Web Services Hypothesis
Recently I hear many investors and analysts pinning their hopes on Amazon Web Services:
Using our estimate of $3.8bn for 2013 AWS revenues, and applying a ~5x multiple based on the comps noted above, we arrive at a valuation of ~$19bn for the business on an EV/Sales basis (equating to ~$41/share of AMZN stock). Importantly, we believe this to be a conservative valuation multiple, as AWS revenues are growing much faster than any of the comps incorporated above. At an 8x valuation multiple, we estimate the AWS business could be worth $30bn as a stand-alone company, or ~$66/share.
I have a lot of experience with both the engineering and the economics sides of cloud computing, and it's a tough business. AWS competes for smaller customers (startups, individual developers) with Google, Microsoft, Salesforce.com (CRM), Rackspace (RAX), and a long list of startups, and for large enterprise customers with all of the previously mentioned companies plus IBM, Hewlett-Packard (HPQ), Oracle (ORCL), SAP, Accenture (ACN), VMWare (VMW), ServiceNow (NOW), etc. These companies offer slightly different products and services, but ultimately they all promise to host and run applications in the cloud.
Rackspace and ServiceNow are the two publicly traded companies that are close to a pure play on cloud computing. RAX trades at forward P/E of 57, which is very rich for a company that grows at 16%. NOW trades at forward P/E of 345. Valuations of cloud services are extremely high on the expectations for surging demand, especially from the enterprise customers.
More importantly for our story is that Rackspace had a net margin of 6% in its most recent quarter, while ServiceNow is losing money. If AWS has margins similar to Rackspace, and Ben Schachter has the correct estimate of its revenues, AWS will be responsible for more than half of all Amazon's profits expected this year.
This makes AWS the only bright spot in Amazon's portfolio, and, by extension, means that the situation in other businesses is even worse. Mr. Schachter expects AWS revenues to grow by 2.3 times over the next two years. Even this explosive growth will only bring AWS revenues to 7% of Amazon's total, contributing 0.42% to Amazon's net margin.
I have serious doubts that AWS will continue to grow at this rate considering the recent developments. This statement warrants a separate article - I would only mention here a huge push by Microsoft to improve its Azure platform this year, the introduction of the PHP runtime on App Engine, the saturation of the mobile app marketplaces, and that the giants of IT services (IBM et al.) were slow to adopt the cloud, but now they take it seriously. From a technology perspective, a year or two ago Amazon's services looked much more attractive relative to the other offerings available at that time. The landscape is more complex and competitive now.
Besides, all major players in this market are involved in a price war. While the demand for the cloud services is growing, prices are dropping, and the total dollar amount is not growing as fast as many analysts assume.
Mark Murphy from Piper Jaffray projects a 44% annual growth in cloud workloads (not to be confused with revenues):
We asked each CIO to describe an application that was moved from on-premise to the public cloud. The cloud winners mentioned most frequently are Salesforce.com (21%), Microsoft (15%), Amazon (9%), Google (6%) and ServiceNow (6%). The losing vendors/technologies most frequently being replaced by a cloud solution are in-house technology (16%), Microsoft (13%), Oracle (6%) and HP (4%).
While the number of applications is a poor proxy for computing loads, and even less so for revenues, it is clear that AWS is not alone in this market and it faces a formidable competition.
The bottom-line here is that AWS is a viable growing business, but it will be a few more years before it is big enough to contribute at least $1 billion to Amazon's profits. By that time investors may start questioning the wisdom of moving all these packages around, streaming movies and selling tablets at a loss, when most profits come from the services division.
The Genius Hypothesis
Every superhero needs magical powers, and the adoring public is ready to bestow them on Mr. Bezos:
When I talk to entrepreneurs in the Valley, our brother up north is the only one who inspires universal fear. No one else in our corner of the world is so good at strategy, tactics, and execution.
I can't quite picture Tim Cook, Larry Page or Mark Zuckerberg trembling at the thought of Jeff Bezos, but I admit that Mr. Bezos is an extremely talented and accomplished individual.
Ironically, if at some point Mr. Bezos is forced out, the shares will probably jump on the hopes of a more profit-minded CEO. I strongly doubt, however, that any other chief executive will have the supernatural ability of Mr. Bezos to keep the company's valuation sky-high.
It is quite possible that Jeff Bezos will move the company in a new direction, giving a new hope for investors to hang on to. As the company grows, however, it becomes more and more difficult to come up with ideas that matter. Goodreads, self-publishing, Amazon-branded cables - these are all nice businesses, but they are too small to make a difference for Amazon's bottom-line.
At some point investors will get tired of getting excited each time a company expands into a new territory. They will demand profits from businesses that the company entered long time ago.
While Amazon.com is overvalued based on every possible metric, its stock will continue to fly high as long as investors continue to believe in the company's profitable future. A weak quarter, especially the very important holiday quarter, may lead to a 10% decline in the stock price, but one bad quarter is unlikely to reverse the upward trend. As long as we are in the bull market, Amazon.com is safe.
Only the continued weak performance of Amazon.com, coupled with a major market correction, will put an end to the love affair that investors have with the company. Otherwise the stock is likely to drift higher.
I recommend that investors with low risk tolerance stay away from Amazon.com.
Investors willing to try their luck on the short side of AMZN ticker may consider opening a position ahead of the upcoming earnings announcement on October 24. A less risky approach is to wait a bit longer. If the third quarter is soft or disappointing, the end of this year may present a good entry point, especially if Santa Claus comes to the Wall Street.
There is a Russian story of Kolobok (doughnut), also known as the Running Pancake or the Gingerbread Man in other cultures, that runs away from different pursuers only to be eaten by a smart fox. I am certain that Amazon.com will escape this fate. Investors would be wise to remember, however, that not every fairy tale ends well.
Additional disclosure: I may initiate a short position in AMZN over the next 72 hours.