User and Subscriber Businesses: The Good, the Bad and the Ugly!

In a series of posts over the course of the last year, I argued that you can value users and subscribers at businesses, using first principles in valuation, and have used the approach to value Uber riders, Amazon Prime members and Spotify & Netflix subscribers. With each iteration, I have learned a few things about user value and ways of distinguishing between user bases that can create substantial value from user bases that not only are incapable of creating value but can actively destroy it. I was reminded of these principles this week, first as I wrote about Walmart's $16 billion bid for 77% of Flipkart, a deal at least partially motivated by shopper numbers, then again as I read a news story about MoviePass and the potential demise of its "too good to be true" model, and finally as I tripped over a LimeBike on my walk home. 

User Based Value

My attempt to build a user-based valuation model was triggered by a comment that I got on a valuation that I had done of Uber about a year ago on my blog. In that post, I approached Uber, as I would any other business, and valued it, based upon aggregated revenues, earnings and cash flows, discounted back at a company-wide cost of capital. I was taken to task for applying an old-economy valuation approach to a new-economy company and was told that that the companies of today derive their value from customers, users and subscribers. While my initial response was that you cannot pay dividends with users, I realized that there was a core truth to the critique and that companies are increasingly building their businesses around their members. 

Consequently, I went back to valuation first principles, where the value of any asset is a function of its cashflows, growth and risks, and adapted that approach to valuing a user or subscriber:

To get from the value of existing users to the value of an entire company, I incorporated the value effect of new users, bringing in the cost of acquiring a new user into the value:

I applied closure by consider all corporate costs that are not directly related to users or subscribers in a corporate cost drag, a drag because it reduces the value of the business:

Cumulating the value of existing and new users, and netting out the corporate cost drag yields the value of operating assets, i.e., the same value that you would derive by discounting the free cash flows to the entire business by its overall cost of capital. You would still need to clean up, by adding in cash, netting out debt and dealing with outstanding options, but that process is the same in both models.

I would hasten to add that a user-based value model is not a panacea to any of the valuation challenges that we face with young, user-based companies. In fact, the difficulties with obtaining the raw data needed on user renewal rates and acquisition costs can be so daunting that any potential advantages that you obtain by looking at user-level value can be drowned out by noise. It is also worth emphasizing that its user-focus notwithstanding, this model is grounded in fundamentals, with value coming, as it always does, from cash flows, growth and risk. I am still learning about this model, but I have put down what I have learned over the last year, when valuing Uber, Amazon Prime and Netflix, into a paper that you can download, read and critique.

Good, Bad and Indifferent User-based Models

One of the motivations for my user-focused valuation was based upon casual empiricism. In my view, many venture capitalists and public investors are pricing user-based companies on user count, with only a few seriously trying to distinguish between good, indifferent and bad user-based models. One of the bonuses of using a user-based model is that it provides a framework for differentiating between great and mediocre user-based companies.

Drivers of Value

A standard critique that old-time value investors have of user-based companies is that they all lose money, but that is not true. There are user-based companies that make money, but it is also true that the user-based model is still in its infancy and that many user-based companies are young, and therefore lose money. That said, there are elements of the cost structure that you can look at, to make judgments on which user-based companies are most likely to grow out of their problems and which ones are just going to grow their problems.

a. Cost Structure: Most young, user-based companies lose money but at the risk of sounding unbalanced, there are good ways to lose money and bad ones, from a value perspective. 

1. Servicing Existing Users versus New User Acquisition: From a value perspective, it is far better for a company to be losing money, because it is spending money trying to acquire new users, than it is to be losing money, because it costs so much to service existing users. The latter signals a bad business model, at least for the moment, whereas the former offers a semblance of hope.
2. Fixed versus Variable Costs: For mature companies with established business models, it is better to have a more flexible cost structure (with more variable costs and less fixed costs). With money-losing, high-growth companies, the reverse is true, since it is the fixed cost portion that yields economies of scale, as the company grows.

b. Growth: Repeating a value nostrum, growth is not always value-creating and not all growth is created equal.

1. Existing versus New Users: A user-based model, where you can grow cash flows from existing users is more valuable, other things remaining equal, than a user-based model that is dependent on adding new users for growth. The reason is simple. Since a company already has expended resources to get existing users, any added revenue it derives from them is more likely to flow directly to the bottom line. Adding new users is more expensive, partly because it costs money to acquire them, but also because new users may not be as active or lucrative as existing ones.
2. Cost of New User Acquisition: This is a corollary of the first proposition, since the value of a new user is net of user acquisition costs. Consequently, user-based companies that are more cost-efficient in adding new users will be worth more than user-based companies that spend considerable amounts on promotion on marketing, to the same end.  

This contrast is best illustrated by looking at Netflix and Spotify, both subscriber-based companies, but with very different models for paying for content. Netflix pays for content as a fixed cost, and derives economies of scale, when it adds fresh subscribers, whereas Spotify pays for content, based upon how much subscribers listen to songs, making it a variable and existing user based cost. As a result, Netflix derives much higher value from both existing and new subscribers:

	Netflix	Spotify
Number of Subscribers	117.6	71
Annual Revenue/Subscriber	$         113.16	$         77.63
Subscriber Service Expenses (as %)	18.90%	79.24%
CAGR in subscriber count	223.93%	369.86%
Value per Existing Subscriber	$         508.89	$       108.65
Cost of acquiring New Subscriber	$         111.01	$         27.30
Value per New Subscriber	$         397.88	$         81.35
Value of all Existing Subscribers	$    59,845.86	$    7,714.28
+ Value of all New Subscribers	$  137,276.49	$  20,764.56
- Corporate Cost Drag	$  111,251.70	$  13,139.75
=Value of Operating Assets	$    85,870.65	$  15,339.10

c. Revenue Models: There are three user-based models, the first is the subscription-based model (that Netflix uses), the second is the advertising-based model (that Yelp uses) and the third is a transaction-based model (that Uber uses). There are companies that use hybrid versions, with Amazon Prime (membership fees and incremental sales) and Spotify (Subscription plus Advertising) being good examples. Each model comes with its pluses and minuses. 

1. Subscription models tend to be stickier (making revenues more predictable) but they offer less upside potential (it is difficult to grow subscription fees at high rates).
2. Advertising models scale up faster, since they require little in capital investment and adding new users is easier (since they free), but revenues are heavily driven by user intensity (how much time you can get users to stay in your ecosystem) and exclusive data (collected in the course of usage).
3. Transaction models are the riskiest, since they require users to use your product or service, but they also offer the most upside, since your upside is less constrained. Amazon Prime's value, in my view, does not stem primarily from the subscription revenues of $99/year but from Amazon's capacity to sell Prime members more products and services.

While no model dominates, picking the wrong revenue model can quickly handicap a business. For instance, using a subscription-based model for a transaction business, where usage varies widely across users, can result in self-selection, where the most intense users choose the subscription-based model to save money, and less intense users stay with a transaction-based model.

Differentiating across User-based Models

With the user-based framework in place, we can start distinguishing between user-based companies. Using existing user value and new customer acquisition costs as the dimensions, we can derive a matrix of companies that go from user-value stars to user-value dogs.

While the combination of high user value with low user acquisition costs may sound like a pipe dream, it is what network benefits and big data, if they exist, promise to deliver. 

• Network benefits refer to the possibility that as you grow bigger, it becomes easier for you to get even bigger, making it less costly to acquire new users. That is the promise of ride sharing, for instance, where as a company gets a larger share of a ride sharing market, both drivers and customers are more likely to switch to it, the former, because they get more customers and the latter, because they find rides more quickly.
• Big data, in a value framework, offers user-based companies an advantage, since what you learn about your users can be used to either sell them more products or services (if you are a transaction-based company), charge them higher premiums (if you are subscription-based) or direct advertising more effectively (if advertising-based). 

Many user-based companies aspire to have network benefits and to use data well, but only a few succeed.

The Pricing Game

As I look at user-based companies, some of which are being priced at billions of dollars, I am struck by how few of them are built to be long term businesses and how many of them are being priced on user numbers and buzz words. Using the framework from the last section, I would like to develop some common features that bad user-businesses seems to share in common and use one high profile examples, MoviePass , to make my case.

Mediocre User-based Companies
Given that so many young companies market themselves, based upon user and subscriber numbers, and that some of them can become valuable companies, are there signs that you can look for that separate the good from the mediocre companies? I think so, and here are a few red flags:

1. All about users, all the time: If the entire sales pitch that a company makes to investors is about its user or subscriber numbers, rather than its operating results (revenues and operating profits/losses), it is a dangerous sign. While large user numbers are a positive, it requires a business model to convert these users into revenues and profits, and that business model will not develop spontaneously. Companies that do not work on developing viable business models go bankrupt with lots of users.
2. Opacity about user data: It is ironic that companies that market themselves to investors, based upon user numbers, are often opaque about key dimensions on users, including renewal (churn) rates, user behavior and side costs related to users. The companies that are most opaque are often the ones that have user models that are not sustainable.
3. Bad business models: If having no business model to convert users to operating results is a bad sign, it is an even worse sign when you have a business model that is designed to deliver losses, not only in its current form, but with no light at the end of the tunnel. That is usually the consequence of having losses that scale up as the company gets bigger, because there are economies of scale. 
4. Loose talk about data: The fall back for many user based companies that cannot defend their business models is that they will find a way to use the data that they will collect from their users to make money in the future (from targeted advertising or additional products and services), without any serious attempt to explain why the data will give them an edge.
5. And externalities: Many user based companies argue that their "innovative" twists on an existing business will both expand and alter the business, leading to benefits for other players in that business, who, in turn, will share their benefits with the user based companies.

The bottom line is simple. It is easy to build user numbers, if you sell a product or service at way below cost, but if your objective is build a long-standing user-based companies, you need a pathway to profitability that is defined early and worked on continuously.

MoviePass: Too Good to be True? 

If you subscribe to MoviePass, for a monthly subscription of $10, you get to watch one theatrical movie, every day, for the entire month. Given that the average price of a theater ticket in the US is $9, this sounds like an insanely good deal, and for an avid movie goer, it is, and the service had two million subscribers in May 2018. MoviePass, though, pays the theaters for the tickets, creating a model that is more designed to drive it into bankruptcy than to deliver profits.

MoviePass Economics

When confronted by the insanity of the business model, Mitch Lowe, the CEO of MoviePass, argued that after an initial burst, where subscribers would see four or five movies a month, they would settle into watching a movie a month, allowing the service to break even. Since Mr. Lowe is a co-founder of Netflix and a former CEO of Redbox, I will concede that he knows a lot more about the movie business than I do, but this is an absurd rationale. If the only way that your service can become viable is if people don't use it very much, it  is not much of a service to begin with.  

In its early days, MoviePass seemed to be trying to build a viable business model, and acquired some high profile venture capital investors, but it was eventually acquired by Helios and Matheson, a data analytics firm, in a  transaction in August 2017. It is Helios and Matheson, intent on giving both data and analysis a bad name, that instituted the $10 a month for a movie-a-day subscription. The subscription worked in delivering users but it, not surprisingly, came with large losses. As MoviePass has continued to burn cash (more than $20 million a month by April 2018), the share price of Helios and Matheson has collapsed, in a belated recognition of its non-viable business model.

Adding to the sense that no one in this company has a grip on reality, Ted Farnsworth, the CEO of Helios and Matheson, argued that the service would continue and had acquired a $300 million line of credit. Since his backing for this line of credit was that he could issue the remaining authorized shares at the current market price, this indicates either extreme ignorance (potential equity issues don't comprise a line of credit) or unalloyed deception, neither of which is a quality that builds trust. Along the way, there have been other attempts to rationalize the model, including the possibility of using the data collected from subscribers to target advertising and the sharing of additional revenues generated by theaters and studios from more movie going. There is nothing exclusive about the data that will be collected from MoviePass subscribers and it is unlikely that theaters and small studios, already on the brink financially, will be willing to share their revenues. In short, this is a bad business model hurtling to a bad end, and the only question is why it took so long.

The Bottom Line

To build a good user-based business, you have to start with the common sense recognition that users are not the end game, but a means to an end. Unfortunately, as long as venture capitalists and investors reward companies with high pricing, based just upon user count, we will encourage the building of bad businesses with lots of users and no pathways to becoming successful businesses.

YouTube Video


Paper on User Based Value

1. Going to Pieces: Valuing Users, Subscribers and Customers

Blog Posts on User-based Value

1. Valuing Uber Riders
2. Valuing Amazon Prime Members
3. Valuing Spotify Subscribers
4. Valuing Netflix Subscribers

Netflix: The Future of Entertainment or House of Cards?

For better or worse, Netflix has changed not just the entertainment business, but also the way that we (the audience) watch television. In the process, it has also enriched its investors, as its market capitalization climbed to $139 billion in March 2018 and even after the market correction for the FANG stocks, its value is substantial enough to make it one of the largest entertainment company in the world. Among the FANG stocks, with their gigantic market capitalizations, it remains the smallest company on both market value and operating metrics, but it has almost as big an impact on our daily lives as its larger peers.

The History

This may come as a surprise to some, but Netflix has been publicly listed for longer than Facebook or Google. The difference between Netflix and these companies is that it’s climb to stardom has taken more time.

Don't get me wrong! Netflix was a very good investment between 2003 and 2009, increasing its market capitalization by 33.36% a year and its market capitalization by about $3 billion, during that period. However, it became a superstar investment between 2010 and 2017, adding about $120 billion in value over the period, translating into an annual price appreciation of more than 50% a year.

The fuel that Netflix has used to increase its market capitalization is its subscriber base, as with the other FANG stocks, the company seems to have found the secret to be able to scale up, as it gets larger. That subscriber base, in turn, has allowed the company to increase its revenues over time, as can be seen in the picture below, summarizing Netflix’s operating metrics.

You can accuse me of over analyzing this chart, but it captures to me the essence of the Netflix success story. While Netflix has been able to grow revenues in each of the three consecutive five-year time periods, 2002-2006, 2007-2012 and 2013-2017, that it has been existence, the company has been faced with challenges during each period, and it has adapted.

1. DVDs in the Mail: In the first five-year period, 2002 through 2006, the company mailed out DVDs and videos to its subscribers, challenging the video rental business, where brick and mortar video rental stores represented the status quo, and Blockbuster was the dominant player. 
2. The Rise of Streaming: It was between 2007 and 2012, where the company came into its own, as it took advantage of changes in technology and in customer preferences. First, as technology evolved to allow for the streaming of movies, Netflix adapted, with a few rough spots, to the new technology, while its brick and mortar competitors imploded. Second, while Netflix saw a drop in revenue growth that was not unexpected, given its larger base, it also saw its content costs rise at a faster rate than revenues, as content providers (the movie studios) starting charging higher prices for content. 
3. The Content Maker: In hindsight, the studios probably wish that they had not squeezed Netflix, because the company reacted by taking more control of its own destiny in the 2013-2017 time period, by shifting to original content, first with television series and later with direct-to-streaming movies. The results have upended the entertainment business, but more critically for Netflix, they show up in a critical statistic. For the first time in its existence, Netflix saw content costs rise at a rate slower than its growth in revenues, with operating profit margins, both before and after R&D reflecting this development. 

The State of the Game

We can debate whether Netflix is a good or a bad investment, but there is no argument that the way movies and television get made has been changed by the company’s practices. It is the rest of the entertainment business that is trying to adapt to the Netflix streaming model, as evidenced by Disney’s acquisition of BAM Media and Fox Entertainment. If I were to summarize where Netflix stands right now, here would be my key points:

1. It's a big spender on content: In 2017, Netflix spent billions on the content that it delivers to its subscribers, and the extent of its spending can be seen in its financial statements. The way that Netflix accounts for its content expenditures does complicate the measurement, since it uses two different accounting standards, one for licensed content and one for productions, but it capitalizes and amortizes both, albeit on different schedules, and based upon viewing patterns. The gap between the accrual (or amortized) cost (shown in the income statement) and the cash spent (shown in the statement of cash flows) on content can be seen in the graph below.

Netflix 10K - 2017

In 2017, Netflix spent almost $9.8 billion on content, though it expensed only $7.7 billion in its income statement. If this divergence continues, and there is no reason to believe that it will not, Netflix’s profits will be more positive than their cash flows by a substantial amount. Note that this divergence should not be taken (necessarily) as a sign of deception or accounting game playing. In fact, if Netflix is being reasonable in its amortization judgments, one way to read the difference of $2.14 billion ($9.8 in cash expenses minus $7.66 billion in accrual expenses) is to view it as the equivalent of capital expenditures at Netflix, since it is expense incurred to attract and keep subscribers.

2. An increasing amount of that spending goes to original content: The decision by Netflix to produce some of its own content in 2013 triggered a shift towards original content that has picked up speed since that year. In 2017, the company spent $6.3 billion on original content, putting it among the top spenders in the entertainment business:

Biggest Spenders on Entertainment Content in 2017

The pace is not letting up. In the first quarter of 2018, Netflix introduced 18 new television series and delivered 12 new seasons of existing series, prodigious output by any studio’s standards. There are three reasons for the Netflix move into the content business.

• The first, referenced in the last section, is to gain more control over content costs and to be less exposed to movie studio price hikes. 
• The second is that Netflix has been using the data that it has on subscriber tastes not only to direct content at them, but to produce new content that is tailored to viewer demands.
• The third is that it introduces stickiness into their business model, a key reason why new subscribers come to the company and why existing subscribers are reluctant to abandon it, even if subscription fees go up.

Netflix has moved beyond television shows to making straight-to-streaming movies, spending $90 million on Bright, a movie that notwithstanding its lackluster reviews, signaled the company’s ambitions to be a major player in the movie business.

3. Netflix has been adept at playing the expectations game: One feature that all of the FANG stocks trade is that rather than let equity research analysts frame their stories and measure their success, they have managed to frame their own stories and make investors and analysts play on their terms. Netflix, for instance, has managed to make the expectations game all about subscriber numbers, and every earnings report of the company is framed around these numbers, with less attention paid to content costs, churn rates and negative cash flows. After its most recent earnings report in January, the stock price surged, as the company reported an increase of 8.3 million in subscribers, well above expectations.

4. The company is globalizing: One consequence of making it a numbers game, which is what Netflix has done by keeping the focus on subscribers, is that you have to go where the numbers are, and for better or worse, that has meant that Netflix has had to go global, with Asia being the mother lode.

At the end of 2017, Netflix had more subscribers outside the US than in the US, and it is bringing its free spending ways and its views on content development to other parts of the world, perhaps bringing Bollywood and Hollywood closer, at least in terms of shared problems.

In summary, Netflix has built a business model of spending immense amounts on content, using that content to attract new subscribers, and then using those new subscribers as its pathway to market value. It is clear that investors have bought into the model, but the model is also one that burns through cash at alarming rates, with no smooth or near term escape hatch.

The Valuation

In keeping with the focus on subscriber numbers that is at the center of the Netflix story, I will value Netflix with the subscriber-based approach that I used to value Spotify a few weeks ago and Uber and Amazon Prime last year.

Cost Breakdown

The key to getting a subscriber-based valuation of Netflix is to first break its overall costs down into (a) costs for servicing existing subscribers, (b) the cost of acquiring new subscribers and (c) a corporate cost that cannot be directly related to either servicing existing subscribers or getting new ones. I started with the Netflix 2017 income statement:

Since Netflix does not break its costs down into my preferred components I made subjective judgments in allocating these costs, treating G&A costs as expenses related to servicing existing subscribers and marketing costs as the costs of acquiring new subscribers. With content costs, I started first with the $2,146 million difference between the cash content cost and expensed content cost and treated it also as part of the cost of acquiring new subscribers. With the expensed content cost of $7,600 million, I assumed that only 20% of these costs are directly related to subscribers and treated that portion as part of the cost of servicing existing subscribers and that the remaining 80% would become part of the corporate cost, in conjunction with the investment in technology and development. One key difference between the Netflix and Spotify cost models is that most of the content costs are fixed corporate costs for Netflix but almost all content costsare variable costs for Spotify, since it pays for content based upon how its subscribers listen to it, rather than as a fixed fee.

Value of an Existing Subscribers

My decision to treat most of the content content costs as a corporate cost has predictable consequences. The costs associated with individual subscribers are only the G&A costs and 20% of content costs, and the number is small, relative to the revenues that Netflix generates per subscriber:

Download spreadsheet

A strength that Netflix has built, perhaps with its original content, is that it has reduced it's churn rate (the loss of existing customers), each year since 2015. In 2017, the annual renewal rate for a Netflix subscription was about 91%, and that number improved even more across the four quarters. In my subscriber-valuation, I have used a 92.5% renewal rate, for the life of a subscriber, assumed to be 15 years. I will assume that Netflix investments in original content will give it the pricing power to increase annual revenue per subscriber (G&A and the 20% of content costs), which was $113.16 in 2017, at 5% a year, while keeping the growth rate in annual expenses per subscriber at the inflation rate of 2%. I estimate after-tax operating income each year, using a global average tax rate of 25%, and discount it back at a 7.95% cost of capital (estimated for Netflix, based upon its business and geographic mix, and debt ratio) to derive a value of $508.89 subscriber and a total value of $59.8 billion for Netflix’s 117.6 million existing subscribers.

Value of New Subscribers
To value a new subscriber, I first estimated the total cost that Netflix spent on adding new subscribers by adding the total marketing costs of $1,278 million to the capitalized portion of the content costs of $2,142 million, and then divided this amount by the gross increase in the number of subscribers (30.84 million) during 2017, to obtain a cost of $111.01 for acquiring a new subscriber. I then net that number out from the value of an existing subscriber to arrive at a value of $397.88/new subscriber right now; I assume that this value will increase at the inflation rate over time.

Download spreadsheet

I assume that Netflix will continue to add new subscribers, adding 15% to its net subscriber rolls, each year for the first five years, and 10% a year for years 6 through 10, before settling into a steady state growth rate of 1% a year. Discounting the value added by new subscribers at a higher cost of capital of 8.5%, reflecting the greater uncertainty associated with new subscribers, yields a total value of $137.3 billion for new subscribers.

The Corporate Drag
The final piece of the puzzle is to bring in the corporate costs that we assumed could not be directly linked with subscriber count. In the case of Netflix, the  technology & development costs and 80% of the expensed content, that we put into this corporate cost category amounted to $6.13 billion in 2017 and the path that these costs follow in the future will determine the value that we attach to the company.

Download spreadsheet

I assume that technology & development costs will grow 5% a year, but it is on the content cost component that I struggled the most to estimate a growth rate. I decided that the accelerated spending that Netflix had in 2017 and continued to have in 2018 reflect Netflix’s attempt to acquire standing in the business, and that while it will continue to spend large amounts on content, the growth rate in this portion of the content costs will drop to 3% a year, for the next 10 years. Note that even with that low growth rate, Netflix will be consistently among the top five spenders in the content business, spending more than $100 billion on original content over the next ten years. Discounting back the after-tax corporate expenses back at the 7.95% cost of capital, yields a corporate cost drag of $111.3 billion.

The Netflix Valuation: The One Number
To value Netflix, I bring together the value of existing and new subscribers and net out the corporate cost drag. I also subtract out the $6.5 billion in debt that the company has outstanding and the value of equity options granted over time to its employees.

Download spreadsheet

The value per share of $172.82 that I estimate for Netflix is well below the stock price of $275, as of April 14, 2018. My value reflects the story that I am telling about Netflix, as a company that is able to grow at double digit rates for the next decade, with high value added with new users, while bringing its content costs under control. I am sure that your views on the company will diverge from mine, and you are welcome to use my Netflix subscriber valuation template to come to your own conclusions. 

It is worth taking a pause, and considering the differences between Netflix and Spotify, both subscription-based business models, that draw their value from immense subscriber bases.

1. By paying for its content, both licensed and original, and using that content to go after subscribers, Netflix has built a more levered business model, where subscribers, both new and existing, have higher marginal value than at Spotify, where content costs are tied to subscribers listening to music. 
2. The Netflix model, which is increasingly built around original rather than licensed content, provides for a stronger competitive edge, which should show up in higher renewal rates and more pricing power, adding to the value per subscriber, both existing and new. 
3. The Netflix model will deliver higher value from subscription growth than the Spotify model, but it comes with a greater downside, because a slackening of that growth will leave Netflix much deeper in the hole, with more negative cash flows, than it would Spotify. 

Now that both companies are listed and traded, it will be interesting to see whether this plays out as much larger market reactions to subscription number surprises, both positive and negative, at Netflix than at Spotify.

In my earlier post on Google, I noted that every company has a value driver, one number that more than any of the others determines value. In the case of Netflix, the key value driver, in my view, is content costs. My value per share is premised not just on high growth in subscribers and continued subscriber value, but also on content costs growing at a much lower rate (of 3%) in the future. To illustrate the sensitivity of value per share to this assumption, I varied the growth rate in content costs and calculated value per share:

To illustrate the dangers to Netflix of letting content costs grow at high rates, note that the company’s equity value becomes negative (i.e., the company goes bankrupt), if content costs grow at high rates, relative to revenue growth, with double digit growth rates creating catastrophic effects. If Netflix is able to cap the costs at 2017 levels in perpetuity, the estimated value per share is approximately $216,  at the base case growth rate of 15%, and if it is able to reduce content costs in absolute terms over time, it is worth even more. In my view, investing in Netflix is less a bet on the company being able to deliver subscriber and/or revenue growth in the future and more one on the future path of content costs at the company.

The Decision

There is no doubt that Netflix has changed the way we watch television and the movies, and it is changing the movie/TV business in significant ways. By competing for talent in the content business, it is pushing up costs for its competitors and with its direct-to-streaming model, putting pressure on movie theaters and distribution. That said, the entertainment business remains a daunting one, because the talent is expensive and unpredictable, and egos run rampant. The history of newcomers who have come into this business with open wallets is that they leave with empty ones. For Netflix to escape this fate, it has to show discipline in controlling content costs, and until I see evidence that it is capable of this discipline, I will remain a subscriber, but not an investor in the company.

YouTube Video

Data Links

1. Valuation of Netflix - April 2018

Blog Posts on Tech Takedown

1. Come easy, go easy: The Tech Takedown!
2. Facebook: Friendless, But Still Formidable!
3. Netflix: The Future of Entertainment or House of Cards?
4. Alphabet Soup: Google is Alpha, but where are the Bets?
5. Amazon: Glimpses of Shoeless Joe!

Spotify Loose Ends: Pricing, User Value and Big Data!

In my last post, I valued Spotify, using information from its prospectus, and promised to come back to cover three loose ends: (1) a pricing of the company to contrast with my intrinsic valuation, (2) a valuation of a Spotify subscriber and, by extension, a subscriber-based valuation of the company, and (3) the value of big data, seen through the prism of what Spotify can learn about its subscribers from their use of its service, and convert to profits.

1. The Pricing of Spotify

I won't bore you by going through the full details of the contrast that I see between pricing an asset and valuing it, since it has been at the heart of so many of my prior posts (like this, this and this). In short, the value of an asset is determined by its expected cash flows and the risk in these cash flows, which you can estimate imprecisely using a discounted cash flow model. The price of an asset is based on what others are paying for similar assets, requiring judgments on what comprises similar.  My last post reflected my attempt to attach an intrinsic value to Spotify, but the pricing questions for Spotify are two fold: the companies that investors in the market will compare it to, to make a pricing judgment, and the metric that they will base the pricing on.

Let's start with the simplest version of pricing, a one-on-one comparison. With Spotify, the two companies that are likeliest to be offered as comparable firms are Pandora, a company that is in the same business (music streaming) as Spotify, deriving its revenues from advertising and subscription, and Netflix, a company that is also subscription-driven, and one that Spotify would like to emulate in terms of market success. Since Spotify and Pandora are reporting operating losses, there are only three metrics that you can scale the pricing of these companies to: the number of subscribers, total revenues and gross profits. I report the numbers for all three companies in the table below, in conjunction with the enterprise values for Pandora and Netflix:

For Pandora and Netflix, the numbers for users and revenues/profits come from their most recent annual reports for the year ending December 31, 2017, and for Spotify, the numbers are from the prospectus covering the same year. To use the numbers to price Spotify, I first estimate pricing multiples for Pandora and Netflix. and then use these multiples on Spotify's metrics:

To illustrate the process, I price Spotify, relative to Pandora and based on subscribers, by first computing the enterprise value/subscriber for Pandora (EV/Subscriber= 1135/74.70 = 15.19). I then multiply this value by Pandora's total subscriber count of 159 million to arrive at a pricing of $2,416 million for Spotify. I repeat this process for Netflix, and then repeat it again with both companies, using revenues and gross profit as my scaling variables. The table of pricing estimates that I get for Spotify explains why those who are bullish on the company will try to avoid comparisons to Pandora and encourage comparisons to Netflix. If, as is rumored, Spotify's equity is priced at between $20 and $25 billion, it will look massively over priced, if compared to Pandora, but be a bargain, relative to Netflix. As you can see, each of these comparisons has problems. Spotify not only has a more subscription-based revenue model than Pandora, yielding higher overall revenues, but its more global presence (than Pandora) has insulated it better from competition from Apple Music. Netflix has an entirely subscription-based model and generates more revenues per subscriber, while facing less intense competition.  The bottom line is that the pricing range for Spotify is wide, because it depends on the company you compare it to, and the metric you base the pricing on. That may come as no surprise for you, but it will explain why there will wide divergences in pricing opinion when the stock first starts to trade, resulting in wild price swings. If you are not adept at the pricing game, and I am not, you should stay with your value judgment, flawed though it might be. I will consequently stick with my intrinsic value estimate for the equity in the company.

2. A Subscriber-Based Valuation of Spotify

Last year, I did a user-based valuation of Uber and used it to understand the dynamics that determine user value and then to value Amazon Prime. That framework can be easily adapted to value Spotify subscribers, both existing and new. To value Spotify's existing subscribers, I started with the base revenue per subscriber and content costs in 2017, made assumptions about growth in each item and used a renewal rate of 94.5%, based again upon 2017 numbers (all in US dollar terms):

Download spreadsheet

Note that revenues/subscriber grow at 3% a year, faster than the growth rate of 1.5%/year in content costs, reducing content costs to 70% of subscriber revenues in year 10, consistent with the assumption I made in the top down valuation in the last post. The value of a premium subscriber, allowing for the churn in subscriptions (only 43% make it through 15 years) and reduced content costs, is $108.65, and the total value of the 71 million premium subscriptions works out to about $7.7 billion.

To estimate the value of new users, I first had to estimate how much Spotify was spending to acquire a new user. To obtain this value, I took the total marketing costs in 2017 (567 million Euros or $700 million) and divided that by the number of new subscribers added in 2017:

Cost of acquiring new user = 700 / (71 - 48*.945) = $27.30
While the number of premium subscribers grew from 48 million to 71 million, I reduced the former value by the churn reported (5.5% of subscribers canceled in 2017). The value of new subscribers then can be computed, assuming that the number of net subscribers grows 25% a year from years 1-5, 10% a year from years 6-10 and 1% a year thereafter (The weakest link in this calculation is the churn rate, which as some of you pointed out is measured in monthly terms. I read this section of the prospectus multiple times to get a better sense of renewal and cancellation rates and here is what I get out of that reading. If the true monthly churn rate is 5.5%, the annual churn rate should more than 50%, meaning that 25 million of the 48 million subscribers that Spotify had at the start of the year left during the year. I don't think that happened, because the total subscribers would not have jumped to 71 million. My guess is that the monthly churn rate reflects how new subscribers become established subscribers, with many trying the service for a month, dropping it, and then coming back again. The annualized churn rate is probably closer to 15%-20% overall and much lower for established Spotify subscribers. I considered using a lower renewal rate in the early years and increasing it in later years, but gave up on it since my information is still hazy. I do believe that will be a key factor in whether Spotify can deliver value, and while the trend lines on the churn rate are good, they need to make their subscribers as sticky as Netflix has made its subscribers.)

Download spreadsheet

In valuing the cash flows from new users, I use a 10% US$ cost of capital, the 75th percentile of global companies, reflecting the higher risk in this component of Spotify's value, and derive a value of about $13.6 billion for new users. (I thank the readers who noticed that I was misestimating my subscriber count, starting in year 2. The numbers should now gel, with the growth rate in net subscribers matching up.)

Spotify does get about 10% of its revenues from advertising, and I will assume that this component of revenue will persist, albeit growing at a lower rate than premium subscription revenues; the revenues will grow 10% a year for the next ten year and content costs attributable to these revenues will also show the same downward trend that they do with premium subscriptions. The value of the advertising revenues is shown to be about $2.9 billion:

Download spreadsheet

The final component of value is mopping up for costs not captured in the pieces above. Specifically, Spotify has R&D and G&A costs that amounted to 660 million Euros in 2017 (about $815 million), which we assume will grow 5% a year for the next 10 years, well below the growth rate of revenues and operating income, reflecting economies of scale. Allowing for the tax savings, and discounting at the median cost of capital (8.5%) for a global company, I derive a value for this cost drag:

Download spreadsheet

The value for Spotify, on a user-based valuation, can then be calculated, adding in the cash balance (1,5091.81 million Euros or $1,864 million) and a cross holding in Tencent Music that I had overlooked in my DCF (valued at 910 million Euros or $1,123 million), and netting out the equity options outstanding (valued at 1344 million Euros or $1660 million):

Download spreadsheet

The operating asset value is slightly lower than the value that I obtained in my top-down DCF (by about a billion), and there are two reasons for the difference. The first is that I did not incorporate the benefits of the losses that Spotify has to carry forward (approximately $1.7 billion) in my subscriber-based valuation, with the resulting lost tax benefit at a 25% tax rate, of about $300 million. The second reason is that I used a composite cost of capital of 9.24% on all cash flows in top down valuation, whereas I used a lower (8.5%) cost of capital for existing users and a higher (10% cost of capital) for new users; that translates into about $600 million in lower value. The value of equity in common stock, the number that will be most directly comparable to market capitalization on the day of the offering, is $19.6 billion.

3. The Big Data Premium?

There is one final component to Spotify's value that I have drawn on only implicitly in my valuations and that is its access to subscriber data. As Spotify adds to its subscriber lists, it is also collecting information on subscriber tastes in music and perhaps even on other dimensions. In an age where big data is often used as a rationale for adding premiums to values across the board, Spotify meets  the requirements for a big data payoff, listed in this post from a while back. It has exclusivity at least on the information it collects from its subscribers on their musical tastes & preferences and it can adapt its products and services to take advantage of this knowledge, perhaps in helping artists create new content and customizing its offerings. That said, I do no feel the urge to add a premium to my estimated value for three reasons:

1. It is counted in the valuations already: In both my top down and user-based valuations, I allow Spotify to grow revenues well beyond what the current music market would support and lower content costs as they do so. That combination, I argued, is a direct result of their data advantages, and adding a premium to my estimated valued seems like double counting.
2. Decreasing Marginal Benefits: The big data argument, even if based on exclusivity and adaptive behavior, starts to lose its power as more and more companies exploit it. As Facebook reviews our social media posts and tailors advertising, Amazon uses Prime to get into our shopping carts and Alexa to track us at home, and uses that data to launch new products and services and Netflix keeps track of the movies/TV that we watch, stop watching and would like to watch, there is not as much of us left to discover and exploit.
3. Data Backlash: Much as we would like to claim victimhood in this process, we (collectively) have been willing participants in a trade, offering technology companies data about our private lives in return for social networks, free shipping and tailored entertainment. This week, we did see perhaps the beginnings of a reassessment of where this has led us, with the savaging of Facebook in the market. 

The big data debate has just begun, and I am not sure how it will end. I personally believe that we are too far gone down this road to go back, but there may be some buyers' remorse that some of us are feeling about having shared too much. If that translates into much stricter regulations on data gathering and a reluctance on our part to share private data, it would be bad news for Spotify, but it would be worse news for Google, Facebook, Netflix and Amazon. Time will tell!

YouTube Video

Data Links

1. Spotify Prospectus
2. Spotify DCF valuation
3. Spotify User-based valuation

Blog Posts

1. Stream On: An IPO Valuation of Spotify

Deconstructing Amazon Prime: Loss Leader or Value Creator?

Update (October 17, 2017): One of the things that I enjoy most about posting my valuations online is the feedback and how much I can use that feedback loop to improve my valuation. There are three changes that I have made to my Amazon valuation, though the end number that I get is not that different. First, as many Prime members outside the US have pointed out, the cost of Amazon Prime is less than $99/year in many countries, ranging from $22/year in Italy to just over $50/year in Germany to only $8/year in India. That lower annual cost will bring down the value of a member (existing and new). To allow for that, I have replaced the $99 annual fee that I had used in my valuation with $93.78, a  weighted average of the fees, allowing for the one quarter of Prime customers in the US who have monthly subscriptions (and pay more) and the 20% of customers outside the US (my estimate), who pay, on average, about $50/year.  Second, as some of you have noted, my operating margin was computed prior to just shipping costs and that I am double counting the customer service and media costs, which should also be added back. That increases my operating margin to 12.11% from 9.19% and I will assume that it improves to 13% over time. Third, and this was entirely my mistake, my value per existing member did not factor in the drop out rate and that has been fixed. 

I am an Amazon Prime member and have been one for a long time, and I am completely hooked. Not only do I (and my family) use Amazon Prime for items ranging from tissue paper to big screen televisions, but it has become my go-to for every possession that I need in my working and personal life. In fact, I know (and am completely at peace with the fact) that it has subtly affected my buying, as I substitute slightly more expensive Prime items for non-Prime equivalents, even when I shop on Amazon. It is not just the absence of shipping costs that draws me to Prime, but the reliability of delivery and the ease of return. In short, it makes shopping painless. As I tally how much we save each year because of Prime and weigh it against the $99 that we pay for it, I am convinced that we are getting far more value from it than what we pay, and that leads to an interesting follow up. If many of the 85 million other Prime members in October 2017 are getting the same bargain that we are, is this not an indication that Amazon has not just under priced Prime, but is perhaps selling it below cost? As someone who has wrestled with valuing Amazon over the last 20 years, I have learned never to under estimate the company. In this post, I would like to take the process I used to value a user at Uber and apply it to value not just a Prime member to Amazon but the collective value of Amazon Prime to the company.

The Growth of Amazon Prime

Amazon introduced Prime in 2005 and the service was slow to take off. At the end of 2011, only about 4% of Amazon customers were Prime members. In the years since, though, the service has seen explosive growth:

In fact, the jump in members in the last three years is particularly impressive, given how much bigger the base has become. In 2016, the company added almost 20 million new members and is on pace to add a similar number this year. In October 2017, the company’s mammoth Prime user base meant that almost 60% of all US households had memberships, suggesting that a non-Prime member is more the exception than the rule for Amazon’s US operations. Growth has been slower outside the US, slowed both by competitive/regulatory pressures and logistical challenges.

The Economics of Amazon Prime

To understand how Amazon Prime works, let’s break down the mechanics. Any one (in the countries where Prime is offered) can become a Prime member, either on a monthly or an annual basis. In the US, in 2017, the annual fee for membership was $99, as it has been for the last few years, and the monthly fee was $10.99. With 85 million members, that translates into a total revenue for the company of over $8.5 billion; the monthly members pay more but there is a portion of the membership (including students) who get discounted memberships. The other benefit for Amazon, though, comes from the fact that Amazon Prime members spend more on Amazon than non-Prime members. While the exact numbers are known only to Amazon, the most recently leaked reports suggest that the typical Prime member spends approximately $1,300/year on Amazon products, as opposed to the $700 spent by a non-Prime member. While it seems obvious, then, that Prime membership leads to more spending ($600, if you believe these two numbers), the statisticians will raise red flags about sampling bias since the true incremental revenue is unobservable; it is the difference between what the existing Prime members are spending ($1300/year) and what those same members would have spent, if they did not have Prime memberships. That is a reasonable point, but there is clearly a Prime impact, where Prime members choose Prime items over less expensive non-Prime offerings on Amazon, just as I do.

The biggest cost, by far, to Amazon is the shipping cost that the company now bears on Prime items. In 2016, the company reported net shipping subsidy costs of $7.2 billion (in the footnotes to the 10K) and assuming that almost all of these costs were related to servicing the 60 million members that Amazon had in 2016 leads to a per-member shipping cost of close to $120/ member. The other free services that Amazon offers its Prime members also create costs, though those costs are embedded in larger company-wide items and are more difficult to separate out. 

• The company expends administrative costs in providing additional customer service to its Prime members. These costs are part of the total G&A cost of $2,432 million that the company reported in 2016, but we will assume that the servicing cost per member was $10 (or $600 million in the aggregate in 2016).
• The company also buys some content for its media and Kindle reading sections specifically to back up its Prime membership. Again, that cost is part of the $16,085 million that the company as its technology and content cost in 2016 and we will assume that 10% of those costs (approximately $1.6 billion) can be attributed to Amazon Prime.

There is one final component of cost that we would like to know, but have to guess at and that is the cost to Amazon of acquiring a new member. That promotional/marketing cost is part of the total marketing cost of $7,233 million that Amazon reported in 2016 and we will assume that this cost is $100/member; in 2016, this would have translated into a total cost of $2 billion to acquire 20 million new members, leaving the remaining $5.2 billion in marketing costs as conventional advertising/marketing cost.  Pulling all these numbers, real and imagined, into a picture, here is what we get as the economics of an Amazon Prime member.

The closing statistic that is worth emphasizing here is that once someone becomes an Amazon Prime member, they tend to stay as members with an annual renewal rate of 96%.

The Value of an Existing Prime Member

Using the numbers from the previous section as a starting point, we are on our way to valuing an existing Prime member. To get to that value, we have to make some estimates for the future that reflect how the base numbers will evolve over time:

1. Renewal Rate: We will assume that annual renewal rates will stay high, at 96%, since the subscription model and the dependence on free shipping makes dropping the service difficult to do. 
2. Incremental Revenue/ Member Growth: As Amazon looks for new products and services to sell its Prime members, we will assume that the company’s legendary marketing skills will work and that the incremental revenue (which we estimated to be $600 and attributed entirely to Prime membership) will grow 10% a year for the next five years. That growth rate will scale down to the inflation rate (1.50%) in year 10, but that cumulated effect will result in incremental sales of $1,275/member in 2027. 
3. Operating Margins on revenues: To estimate the operating margin on revenues, I started with Amazon's operating margin but then added back the shipping, customer service and Prime media acquisition costs, since I treat them as separate costs. The resulting margin is 12.11%.
4. Shipping Costs: The biggest cost to Amazon is shipping and much of what the company seems to be doing both in terms of new investments (in distribution centers, trucks and drop off locations) and acquisitions (Whole Foods) seems to be designed to keep these costs in check. We will assume that shipping costs will grow 3% a year for the next five years (well below the incremental revenue growth), before settling into growing at the inflation rate thereafter. 
5. Customer Service Costs: The cost of Prime member customer service will increase 5% a year for the next 5 years and the inflation rate thereafter.
6. Risk and Cost of Capital: I will assume that Amazon’s overall cost of capital applies as the right risk adjusted rate to use on all of its member valuations, since they partake in its entire product line. That cost of capital, in October 2017, given a US treasury bond rate of 2.35% was 8.00%.

With those assumptions in place, we can estimate the value of a Prime member:

Download spreadsheet

With our estimates, the value per prime member is approximately $486 and the total value of 85 million prime members is $41.3 billion.  While you may view this value as built on a mountain of guesstimates, and you would be right, the analysis does provide guidance on the drivers of Prime member value. The two biggest are:

1. Growth in incremental revenues: Getting Amazon Prime members to buy more products and services is key to their value and it should be no surprise to see stories like this one. As revenue growth climbs from 10% to 15%, for instance, the value of an existing member becomes $744 and the value of member base increases almost 65%.
2. Keeping shipping costs contained: The key to extracting value from Prime members is checking shipping costs. To illustrate, if shipping costs grow at the same rate as incremental revenues do, the value per member collapses to $71.50. 

Viewing Prime members through this prism makes it easier to explain the Whole Foods acquisition, by Amazon, for about $14 billion. Rather than think of it as Amazon’s entrée into a low-margin, intensely competitive grocery business, it would make more sense to view it as an acquisition of a distribution system (of 460 Whole Foods stores, in prime locations) that will reduce shipping costs in the future, while also providing a new menu of products/services that can be offered to Prime members. That is bad news for a whole host of other players in the market but that is a story for another day.

The Value of a New Member

To get from the value of an existing member to that of a new member, you need to have a measure of how much Amazon is spending to acquire new members. As I noted earlier, the company is opaque on this issue, though I would hazard a guess that it is a much more onerous number outside the United States. If you work with my guess of $100/new member as the base year number, we need only one more estimate to get to the value of new members and that is the growth in the membership base. Given the success that the company has shown on this front in the last five years, we will assume a growth rate of 15% for the next five years (which will bring Prime membership to 155 million in 2022). Given that large base, we will scale growth down to 5% a year from years 6-10 and to 2.25% a year thereafter.

Download spreadsheet

Again, with our estimates, the value of new Prime Members is approximately $53.9 billion and that number, in addition to being sensitive to our estimates of growth in members, magnifies the effects of incremental revenue and shipping cost growth that affected the value of an existing member. Thus, setting shipping costs to grow at the same rate as incremental revenues makes the value of new members a negative number, suggesting that growth will become value destructive if shipping costs are not brought under control.

The Corporate Drag

While it is tempting to stop and add the value of existing members and new members to arrive at the value of Amazon Prime, you would be missing a significant cost that we will term the corporate drag. To feed its ambitions with Prime, Amazon is spending far more on media content (books, movies, TV shows) than it would otherwise have and those costs will continue to grow with Prime. Earlier, we assume that 10% of the company’s current technology/media costs are attributable to Prime, yielding a base year cost of $1,609 million. We will assume that these costs will grow with the number of Prime members, yielding the following value for future costs:

Download spreadsheet

In effect, we will lower the value of Amazon Prime by $32.8 billion to reflect these additional costs. Note that though I treat this cost as a drag, it is not wasted, since it is the additional media that is being offered as a sweetener for existing Prime members to stay on and new ones to join.

The Value of Amazon Prime

Now that we have valued Amazon Prime’s existing members, its new members and the corporate drag, it is a matter of bringing them all together into a consolidated value. In our judgment, Amazon Prime is worth $62.2 billion to Amazon:   

Category	Value/Cost today	Details
Value of Existing Members	$41,334.69	Value of 85 million members @$486/each
Value of New Members	$52,792.78	Value of new members added
- PV of Corporate Expenses	$32,845.63	Value of additional media/tecnology costs
Value of Prime Membership	$61,281.83	Overall value of Amazon Prime

I know that I have made a multitude of assumptions along the way to get to this value and that you may disagree with many of them. As always, you can download my spreadsheet and make the changes that you think need to be made. If you work at Amazon or view yourself as an expert on Amazon (I am not), your numbers should be much better than mine, and I would hope that your valuation will reflect the better information. While I have made a few optimistic assumptions to get to the value of Amazon Prime, I believe that there is an additional value that I have not counted in. Amazon is building a base of loyal, intense members that it can draw on to promote whatever its next product or service is, whether it be in retailing, technology, entertainment or cloud computing. That value is what you could call a real option, though those words are used far too frequently in places where they should not be, and that real option may be Amazon’s ultimate wild card.

Conclusion

I have long described Amazon as a Field of Dreams company, one that goes for higher revenues first and then thinks about ways of converting those revenues into profits; if you build it, they will come. In coining this description, I am not being derisive but arguing that the market's willingness to be patient with the company is largely a the result of the consistency with Jeff Bezos has told the same story for the company, since 1997, and acted in accordance with it. Amazon Prime symbolizes how Amazon plays the long game, an investment that has taken a decade to bear fruit, but one that will be the foundation on which Amazon launches into new businesses. I know that there are many companies that model themselves after Amazon, but unless these "Amazon Wannabes" can match its narrative consistency and long time horizon, it will remain a one of a kind.

YouTube Video


Attachments

1. Amazon Prime Valuation
2. Amazon 10K (2016)

Previous (related) blog posts

1. User/Subscriber Economics: The Value of an Uber User (June 2017)
2. User/Subscriber Economics: Value Dynamics (July 2017)