It's important to master the simple before you can understand the complex. I'm not sure if that's a quote from a famous figure or just something that my history teacher made up, but for some reason it has always stuck with me. It seems that no matter the application, it rings true. In the case of finance and valuations it is especially important. The three "technology" stocks I follow and dislike the most are: Apple (AAPL), Amazon (AMZN), and Netflix (NFLX) because all three have (or had) valuations that I considered to be ridiculous.
The most basic valuation principle, and probably the easiest to understand, is the price earnings ratio (P/E). You simply take the market cap of a stock and divide it by the earnings for a period of time (either historical or future). Analysts usually determine the P/E multiple a stock deserves by looking at something like the S&P 500 and analyzing its P/E multiple, then looking at the company they are trying to value and determining the growth prospects. If growth in earnings is estimated to be higher than the S&P 500 (or some analysts do sectors or a direct comparable) then the company is usually given a premium multiple and conversely if it is lower then it is given a lower multiple. Pretty simple and pretty easy to master… right? Unfortunately, the answer is no. Even though P/E is very simple to calculate, its nuance is much harder to grasp.
What does P/E really mean? Well in mathematical terms it is actual the number of periods (generally years) that it would take for the company in question to earn its current valuation and theoretically pay back your investment. So if you think about it in those terms a second, more important, aspect to P/E multiple valuation comes up: the ability of the company to generate those earnings over that time period. While this is, in some ways, tied to growth it is also tied to the industry they are in and the product they are selling. It is entirely reasonable to assume that we will need soap for the foreseeable future, so in many ways that means companies like Johnson and Johnson (JNJ) deserve a premium valuation (in terms of P/E multiple) when compared to a company like AAPL (that has/had superior growth).
In effect, this is what analysts frequently get wrong when they claim a company is undervalued. When AAPL was at its peak and analysts were placing $1,000 price targets on the company, the most familiar refrain was "the P/E multiple is low relative to historical and the market". The reality is the market was discounting the fact that we have absolutely no way of knowing what kind of profit AAPL will generate five years from now, let alone 10 or 14. In all likelihood the smartphone market we know today will be completely different from the one we know 5 years from now, and in all likelihood AAPL will not be the leader in that market. So why should investors be willing to pay 10 times earnings (or even market average P/E) when the future of its product, let alone earnings, is about as clear as mud.
AMZN is a different animal, in that I am actually quite confident that their company and service will be around for the rest of my lifetime (barring gross mismanagement), meaning it should receive a premium multiple. The problem is, the company doesn't really make very much money in terms of operating income (and therefore earnings per share). According to company filings, it has gone from $24 billion in revenue in 2009 to $61 billion in revenue for 2012. Amazon's operating income actually declined from $1.1 billion to $0.7 billion on a 2.5x increase in revenue. Even if we just assume that 2012 was a bad year (although its operating income has been declining or staying effectively flat since 2009 on materially increased revenue) and it will return to generating profits in the future, how big will those profits be? AMZN competes on price over the Internet. To check an alternate price it takes just a few key presses and the barriers to entry are extremely low. It is also reliant (almost entirely) on products provided by 3rd parties (i.e. if you buy a book it isn't necessarily published by AMZN, or a movie, or a TV, etc). Ultimately this means it has no real ability to increase margins in the future as it will always be competing on price. So with a 3-5% operating margin (high by their current standards), it would need to sell literally hundreds of billions in product to generate enough EPS to get to $15 in earnings (20 x $15 = $300 its current price, 20 being the premium valuation companies like JNJ and AMZN should receive because we know they will be around forever). For the sake of argument that's a 5x increase in revenues just to get current shareholders to breakeven, not a bet that I would make.
NFLX, is what you get when you combine the worst attributes of AAPL and AMZN. Below are my top reasons why the NFLX valuation is ridiculous, even though my (overly optimistic) model says it could generate EPS of $0.70 in the quarter it is reporting tonight.
- The company doesn't make any money on its streaming business currently. If you assume a revenue attribution model for its "other operating expenses", then DVD will provide about $100mm in contribution profit this quarter and $70mm in net profit. Streaming will provide $105mm in contribution profit and around $0 in net profit. This is on an estimated 37mm paid streaming subscribers.
- The streaming business growth is the reason for the crazy multiple. We already know it doesn't currently make (much) money. However, like AMZN, NFLX primarily just provides other people's content to subscribers at a cheap price. The cheap price is why they have so many subscribers and such strong subscriber growth. If they reduce content spend or increase price it's likely growth in subscribers will stop or reverse. So in terms of share price they are stuck in a cycle of having to spend to grow and having to grow to justify valuation all while the number of available subscribers is shrinking (mostly because they have so many and keep adding more, a similar problem that AAPL faced).
- The streaming business is also a complete unknown. I can guarantee that 10 years from now the way we watch TV or movies at home will be different than the way we do so today. Whether the NFLX "all you can eat" streaming content model is the way things go isn't clear to me. It's possible we all cut the cord and several dozen "channels" (like NBC, etc) all create their own streaming services (most already have it, they just limit the amount of content that is available) or maybe NFLX's current content providers will do what Time Warner (TWX) did and create their own streaming service. Any way you cut it, it's likely that the competition for subscribers will get intense in the future and NFLX will only be able to compete on content by paying increasingly higher prices.
- Creating its own content is smart… but the reason for the premium valuation isn't because House of Cards won some Emmy awards. Creating and selling entertainment content is an old business and goes back to the days before TV. It isn't an exciting or sexy business and all the hype around how critically acclaimed NFLX shows are makes no sense. Most struggle to make money (DWA) and/or trade at 20x P/E or less.
- The valuation is crazy. As noted above there are four (GOOD) reasons why the valuation is crazy, but there is just one great one: the content spend is relatively low in comparison to its market cap. When Carl Icahn bought shares he thought that NFLX would be a good take-out target. At today's valuation that is no longer true. Why would GOOG or AMZN pay $15bn (likely $20 to $25bn if they had to pay a premium to get shareholders to agree) for NFLX when they could just spend $2bn on content? Because NFLX doesn't own anything other than temporary exclusive rights to streaming content, a potential acquirer could easily just "test" the waters by creating their own service. Let's pretend GOOG wants to get into the streaming business. They already have a distribution network in the form of YouTube, so all they would need to do is offer YouTube premium for some monthly fee. If they buy content from a provider at a rate of a couple billion a year (just like NFLX spends) then they would have, in effect, recreated NFLX at an extremely low price. In fact, they would have 7 to 8 years to fiddle with the service and get it perfect before they even approached what the NFLX purchase price would be.
As an exercise I tried to figure out what would need to happen to have the current NFLX valuation be "fair". I used 15x P/E as the reasonable valuation number, meaning the company would need to generate around $1bn in annual profits after tax. The reason for this seemingly "average" P/E multiple, is illustrated above. NFLX has a weird combination of the strengths and weaknesses of AAPL and AMZN. On the "premium" valuation side it has strong earnings growth (albeit from nearly zero recently) and strong revenue growth. On the "discount" valuation side it is in a business that will soon be hyper-competitive, it competes on price meaning margins will be low, it has little to no control over content and its costs, and it is approaching market saturation in its biggest market (the US). I used 15x to be optimistic, even though it probably deserves a 10x P/E.
In order to estimate how NFLX could generate $1bn in annual profits after tax I used two methods. One is stripping out the DVD business entirely, holding all expenses constant and changing the number of subscribers. This method isn't really realistic, in that I don't think NFLX could add so many subscribers without increasing content spend and general expenses, but these optimistic assumptions resulted in a number of 55mm paid subscribers. To make today's valuation fair they need to add 21mm (61%) subscribers (using method #1). The other method was to leave DVD in and look at contribution profit gains per percentage gains in total streaming subscribers. This method has its problems as well, but based on last quarter (arguably their best) a 13% increase in subscribers equates to a 23% increase in contribution profit. This method (also very aggressive and optimistic) means in about 5 quarters (at 13% subscriber growth) annual profits would be about $1bn, which equates to 63mm paid subscribers required to make today's valuation fair (using method #2).
The problem for NFLX longs (and shorts) is that those numbers, while probably low, are theoretically achievable (probably not in the time frame mentioned). It isn't crazy to think that NFLX could end up with 60 to 70mm subscribers. What's crazy is that is the breakeven for current NFLX investors, so you're investing in the company hoping that everything goes perfectly just so you will have a chance to make a very low return (possibly zero). If things don't go perfectly you stand to lose a substantial portion of your investment. That's why I don't understand why anyone remains long at these levels unless red is, in fact, the new black.
Based on the above points I suspect that NFLX will report a quarter (maybe not this one, but within the next four) where they report decent subscriber growth and ok growth in earnings but the stock will either not react or go down. That is exactly what happened to AAPL and will be the signal that the market might finally be looking to place a reasonable valuation on NFLX's current and future earnings. My (overly optimistic) estimates say they might report up to $0.70 per share in earnings on 37mm paid subscribers tonight. That figure is well above the current street estimate of around $0.40, so there is a chance that my short position (through options) may not work out. I am confident, however, that either myself or NFLX longs will be saying "I've made a huge mistake" by the time the dust settles tomorrow.
Additional disclosure: Short through long puts.