While the company is still very much in the accelerating phase of growth, not all is well..
1) Yes, subscriber growth is accelerating:

• 2007: +1.2M
• 2008: +1.9M
• 2009: +2.9M
• 2010: +7.7M

2) And yes, this is producing three virtuous cycles (according to CEO Hastings)

1. More subscribers means more money to license content with, which drives more subscriber growth
2. More subscribers means more word-of-mouth from subscribers to those who are not yet subscribers, which drives more subscriber growth
3. More subscribers means we can increase R&D spend to improve our user experience, which drives more subscriber growth

3) And yes, Netflix has still plenty of room to grow, both in the US, but also now in Canada and in other countries abroad (planned for the second half of this year). Although that growth will once stop, the US has some 70M households with internet connection capable of streaming. a full 40% of all households don’t have high-speed internet connection (but perhaps some of them will in the future..)

4) Further, there are opportunities with new devices like smartphones, tablet computers, and the like. From the letter to shareholders accompanying the Q42010 results:

There are now hundreds of CE device SKUs that support streaming from Netflix, and our efforts are turning to rapidly improving our user interface on all of these devices. Some of the improvements that we’ve recently developed, and tested in controlled trials, have measurably increased viewing and retention, which are our two best indicators of satisfaction. We have a long list of promising concepts that we can’t wait to test over the coming years. Some of these are purely visual, such as optimal content graphic size, but most are algorithmic, driving a very specific personalized set of suggestions for each subscriber.

The devices with large installed bases – meaning Windows and Mac laptops, Sony PS3, Microsoft Xbox, and Nintendo Wii – are the most popular devices for watching instantly from Netflix. AppleTV has done very well for us, and in just four months has passed the also-growing iPad in Netflix viewing hours. The Roku player remains a strong performer. Our iPhone and Windows phone applications are very popular, and we’ll have support for several Android phones this year. While Google TV has not yet gone mainstream, the concept of Android and Chrome built into televisions and Blu-ray players is powerful;we’re confident that they will be very successful, and we are investing in our Google TV application. Blu-ray players and TVs with Internet connectivity are the big growth categories for us, as more and more providers are building in Wi-Fi across their product lines.

Our partners are finding that Netflix is their most popular service for Internet video viewing, and they are seeking ways to make their products easier to use. At the Consumer Electronics Show a few weeks ago, we announced the launch of our “one-click access” program where CE device remote controls have a red Netflix button providing convenient, direct access to Netflix for consumers. Initial partners include Samsung, Sony, Toshiba, Roku, and Boxee, with more coming.

However…

5) Amazon might start competing directly, Engadget was able to catch a few experimental screenshots that had text like this:

“Your Amazon Prime membership now includes unlimited, commerical-free instant streaming of 5,000 movies and TV shows at no additional cost“

6) Average monthly revenue per user (“AMRPU”)

has come down in nearly sequential fashion during this period, from $14.09/month in Q1 2008 to $11.64 in Q4 2010. This, I would guess, is primarily the result of the company passing along cost savings from their increased economies of scale coupled with the mix changing to more streaming subscriptions. [Jay Schembs]

This is not a disaster (not yet anyway), as

The benefit from accelerating subscriber growth has far outstripped the effect of sequential declines in AMRPU, resulting in revenue growth compounding at nearly 22% per year between FY2007 and FY2010 to $2.16 BN in FY2010.[Jay Schembs]

6) But it doesn’t generate much free cash-flow

NFLX’s ability to generate free cash flow (FCF) is severely restricted by the significant ongoing capital expenditures required to secure content. As can be seen in the exhibit below, any operating leverage analysts might expect in the business is limited by content capex. Indeed, free cash as measured by the company was actually negative in Q3 2010 as streaming content capex began to significantly increase. Q4 2010 free cash was positive even as these content acquisition costs continued to increase as a proportion of revenue; however, much of Q4’s free cash resulted from changes in balance sheet accruals.

And with some redrawing of the figures, the conclusions are not too friendly…

To help focus on true operating performance, I have made certain adjustment to the company’s cash flow statements. First, the company for some reason lists streaming content acquisition costs as a component of operating cash flow, yet discloses DVD content acquisition costs as an investing cash flow. For consistency (and, in my mind, as a matter of clearer disclosure), I have pushed streaming content acquisition costs to investing cash flows. Second, I have stripped out the accrual component of operating cash flows. The result is what I call adjusted operating cash flow (adjusted CFO), which as you can see below has been remarkably consistent at slightly above 20% of revenue over the last three years.

When gross capex (resulting from DVD content acquisition, streaming content acquisition and traditional capex) is netted against adjusted CFO, I believe we can more clearly see operating FCF. As the below chart illustrates, adjusted FCF has historically hovered between 0-10%, but began to move significantly negative in Q3 2010 and continued in Q4 2010. This downward move in adjusted FCF margins is directly coincident with sharp increases in content capex as a percentage of revenue (as seen in the above exhibit).

So we have a business growing rapidly on the top line that requires substantial ongoing capital investment. What are reasonable assumptions for the future? Rather than identify specific inputs to generate a singular value, I prefer to look at a range of reasonable outcomes and consider the values implied by those outcomes.

By way of comparison, revenues have compounded at 25% since FY2008, and FCF margins (not adjusted as above) have averaged 6%.

Assuming 5% terminal growth and 11% cost of capital (both optimistic assumptions in my mind), a 5-year discounted cash flow model generates the following range of enterprise values, all in comparison to a current share price of $218.

Everything I have discussed thus far relates to operations. My final comments relate to capitalization and management. The company has spent nearly $750 MM buying back stock over the last three years, much of which they’ve simply purchased from selling insiders. I don’t recall seeing an S-4 report as ugly as NFLX (see here). Further, long time CFO Barry McCarthy stepped down in December. While he may have been seeking a CEO role, he likely had no chance with NFLX. This can’t be construed in any way other than a negative for the company.

As I said in my initial comments, I welcome critique and hope this sparks substantive debate.

Disclosure: I am short NFLX.