I'm an avid reader of SA, and along with screening for new ideas, I'm also always on the lookout for views opposing my positions, hoping to learn something new or to revise my analyses. Recently in my readings, I've been struck by the bullish exuberance and insouciance towards potential risks that leads many authors to project optimistic assumptions well into the future. To me this is a symptom of today's raging bull market and if this approach is indicative of even a minority of longs, I can't help but think that many bulls will be sorely disappointed in their company's actual future results.
For instance, I'm currently short NFLX (with nothing but losses to show for it so far), so it was with great interest that I clicked on a recent SA article predicting that Netflix (NASDAQ:NFLX) might achieve $35 billion in revenue by 2024, and a market cap of $140 to $175 billion to go with it. Upon reading the article, however, I realized that it was the epitome of the type of attitude and approach that's become so pervasive lately. As such I've chosen it as the foil for my commentary, but rest assured that you can find many similar examples throughout today's financial press.
I'm unaware of any script or outline for these bullish analyses, but it seems that they often begin by discussing some variant of the "market opportunity" just waiting to be seized. In many cases, this is done by citing the total addressable market (TAM). In this particular article, it's by reference to an observation about consumption demographics (which purportedly reflects on the TAM). The purpose is always to highlight some huge number which will make all the other numbers in the analysis seem reasonable and conservative by comparison.
Beginning an analysis by reference to a TAM is strictly a raging bull market phenomenon because - other than a few blockbuster or orphan drugs that have no real competition - the TAM is largely irrelevant in terms of an individual company's prospects. This is in part because: (1) competition typically also sees the TAM and gears up accordingly; (2) execution is critical and few companies can easily scale to the extent warranted by the ephemeral TAM; (3) the TAM is almost always defined so widely that it doesn't really tell you much about the potential of any one company's narrower offering within the wider market.
As a thought-experiment in this regard, imagine if we valued insurance companies or steel manufacturers or micro-breweries or taxi cab companies by beginning with their theoretical TAM. Here most people would think it's silly, but during a frothy bull market, tech stocks have their own (temporary) valuation rules and models.
In this particular case, we learn that millennials are almost three times as likely to watch TV online, and by that stat alone, Netflix is in the driver's seat. (Note that there's no discussion of competition; nor of the economic relationship between content creators and content distributors; nor of the potential for infrastructure providers to impact the market; nor of all the potential changes in viewing habits that may occur during the next 10 years). Instead one graph tells the whole story of the vast opportunity of seemingly low-hanging fruit for Netflix to waltz in and take advantage of.
Next, we have 10-year (!) projections of monotonic and consistent growth, both of subscribers and of pricing increases. The trick here is to start with a model that might be "a little too aggressive" and then make it reasonable by cutting it in half.
Because everyone knows that half of "a little too aggressive" is conservative.
In this fashion, the new model estimates NFLX's ten-year subscriber growth at 12.67% compounded annually accompanied by pricing increases of 9.56% also compounded annually. These are said to be extremely attainable figures. Why? Well there's really no argument for it, other than it's only half of an assumed rate that's "a little too aggressive". But in a Fed-driven bull market, apparently that's enough.
Slap on a 20% net profit margin and the analysis is done, NFLX is woefully undervalued. (Of course there are token warnings about volatility, and for some reason a statement about NFLX being "priced for perfection", but the real verdict is that NFLX is a buy due to the beauty of compounding.)
I've already mentioned the absence of any discussion about the mechanics of the business or the impact of competition, but let me also note the absence of any macro economic risks. Instead we have a blasé linear projection over 10 years with no regard for business cycles, potential impact of rising inflation and/or interest rates, etc. Again, this type of approach is only possible in a raging bull market, and, in my view at least, is a huge setup for massive disappointment down the road (whether that be in a year or in four).
The main point of this article is to highlight the reckless optimism that now pervades the stock market community - an optimism that leads to naïve valuation models - and I think my general commentary above achieves this. But to give it a little more color and depth, let's look a little more closely at the assumptions used to predict NFLX's 2024 market cap of ~$157.5 billion. (A market cap which, assuming no change in share count - another factor which is omitted from the analysis - results in a share price of ~$2,600.)
First, let's look at the projected subscriber growth and price increase data which yield the revenue number. The "conservative" projections here are 12.67% and 9.56%, respectively, which together result in a revenue compounded annual growth rate "CAGR" of 23.4%. There are a few data points mentioned in the article which might be used to support these figures:
- Over the past five years Netflix has grown revenues by 160%
- Netflix will eventually increase subscriber prices by $1 or $2 says its chairman
- Year-over-year Netflix has grown its subscriber base by 35%
Remember that the author is engaged in projecting a 10-year growth rate, so one would think that the previous five-year revenue growth rate, having a much smaller initial base, would at least be relevant to "sanity check" the future projection. But there's no mention of the rate. So for what it's worth, a 160% increase over five years equates to a CAGR of 21%. This is less than the forward ten-year projected revenue growth rate.
Note too that the year-over-year 35% subscriber base growth is mostly contained within the 5-year revenue growth rate already considered, which, at least to me, speaks loudly to the type of volatility that's likely with NFLX's revenue growth rate going forward. In other words, the 35% YoY number is a cautionary signal because it means that there must have been much smaller growth years in the other 4 years for NFLX to have only a 21% revenue CAGR over 5 years. This in itself would show the audacity of pretending to project constant numbers forward for 10 years, much less using a value of 23.4%.
Another way to ground projections is to realize that NFLX opens new international markets from time to time, and these can cause a big one time jump in revenues. But for a ten-year growth rate to hold, the important number will be organic growth in mature markets. We have an excellent way to ballpark that because in the third quarter of 2011 NFLX began breaking out paid domestic streaming and DVD members (at the end of period). At a minimum, the growth rate of total domestic NFLX members should be a valuable data point for future projections. (And note that here it's important to use the total subscribers because during the period NFLX was actively seeking to migrate DVD subscribers to streaming, hence the streaming subscriber numbers on their own don't tell the real new subscriber growth story.)
At the end of September 2011, NFLX had 34,324,00 paid domestic subscribers. At the end of the first quarter of 2014 (the latest available numbers), it had 40,866,000 paid domestic subscribers. If I've done the math right, that's a quarterly subscriber rate of 1.76% and an annualized rate of 7.24%. That's almost half of the forward subscriber growth used in the bull market NFLX analysis. (Of course, if you don't look at the historic numbers, they won't cause you to doubt your conclusions.)
Next, let's consider the projected price increases. Here's a graph of the quarterly revenue per subscriber for the data NFLX handily breaks out. (And as an aside, NFLX is very helpful in providing easy to read letters to shareholders and financials in Excel format on its website. No need to manually enter data from 10-Qs and 10-Ks on the SEC's website. But they're not much help if analysts choose not to look at them…)
The one thing that jumps out from the graph is that there have been NO price increases during NFLX's history so far (and the one that was proposed was rescinded in what amounted to a fiasco for shareholders). So to project annual price increases of 9.56% coupled with much higher than trend subscriber growth seems ludicrous, but in a bull market any assumption and projection is apparently sound.
A generous but realistic price increase assumption might be $1 per month every 5 years, which would take the $23.23 per quarter number to $27.23, for a 10-year CAGR of 1.6%.
Finally, let's examine the basis for asserting 20% net margins. Again, there's no argumentation in the article, because in today's market the burden of proof (i.e. of examining assumptions) falls squarely on the shoulders of bears. So let's pick up this asymmetric burden.
The chart below plots NFLX's historic net margins. Nary a 20% in sight. But of course bulls will argue that that's due to growth. (How this squares with our author projecting higher than trend subscriber and price growth along with 20% net margins is probably a secret only bulls are privy to.)
NFLX Profit Margin (Quarterly) data by YCharts
To compensate for growth, let's again turn to the mature domestic market for guidance. By looking at the numbers, one might guess that perhaps our venerable analyst substituted NFLX's preferred metric of contribution margin for net margin? After all, they're both margins, and NFLX puts the contribution margin front and center in its press releases. And indeed in the latest quarter, domestic streaming has a contribution margin of (201,265 / 798,617) = 25.2%, so maybe this was used to conservatively(!) ground the estimate?
But what exactly is contribution margin? Here's NFLX's definition (from the latest 10-Q):
So contribution margin doesn't account for:
- Technology and Development costs
- General and Administrative costs
- Interest Expenses
- Income Taxes
In other words, only a very optimistic bull could substitute contribution margins for net margins.
Since NFLX doesn't give us net margins per segment, let's build a proxy for it in the mature domestic streaming business by adding back those four expenses pro rated by the domestic streaming revenues to total revenues. (This is probably too low, since all of the income taxes are likely due to the domestic segment.)
The graph below shows the results:
The maximum segment net margins are thus 10%, but - with the rapid advance of competition and the resulting improvement of content creators' negotiating position along with continued international losses and higher tax burdens - a 5% margin, which is slightly better than the historical profit margin of 4%, is probably the better forward estimate.
The table below summarizes my historically-grounded rate estimates - which are all off of smaller bases, and at a time when NFLX's chief competitor was Blockbuster (not Amazon, Google or the cable companies) - compared to our intrepid bull market analyst's numbers:
Based on this, and all the risk factors not even considered here, I see no reason to revise my bearish outlook on NFLX... however much this may hurt temporarily.
Disclosure: The author is short NFLX. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I am a Netflix DVD subscriber. I actively trade around core positions