I’ll state at the outset that I’m neither bearish nor bullish on Netflix Inc. (NFLX), but I’m certain that I will be one of these by the time I finish my three analyses on the company. In this, the first of my analyses, I’ll take a “top-down” approach to answer two critically important questions. First, what is the size of the global market for Netflix? This is critical, as the company itself suggests that its future depends on with international expansion. Second, I want to look at the assumptions about future growth embedded in the current stock price. Marrying these two questions will, in my view, go a long way in determining whether the assumptions about future growth are reasonable. In my second study of the company, I’ll get more into the weeds and try to work out how extraordinary (or not) this company is. It may be that nervousness about valuations is unwarranted. Only a dive into the weeds can offer insight on that subject. My final analysis on Netflix will look at changes in how the market reacts to similar news. If the shares jumped 20% on a surprise 3 years ago but only 2% this year on a similar surprise, I think that offers us some insight into the changing “mood” of the market for this company’s stock.
My impression about analysis of Netflix, in particular, is that we’re very much prone to bias of one sort or another, and that’s what I’m explicitly trying to avoid in my pieces on this company. We tend to be either excessively optimistic or excessively pessimistic about Netflix’s future, and those pre-judgements taint our analyses. Further, we tend to be a bit overconfident about what we just “know” about the future of this business. There’s a rich literature on the problems associated with these two behavioral biases. For my part, I think these biases are so destructive to shareholder wealth that I co-produced a film on behavioural finance where we interviewed some of academic the giants in the field (Dan Ariely, Shlomo Benartzi, Hersh Shefrin, Richard Thaler, Terry Odean and others). The message from these academics is fairly plain: overconfidence and overoptimism cost money. Finally, I don’t think I’m exaggerating when I suggest that people pick apart every piece of data that comes out of the company in the hope of gaining some insight that no one else will spot (there’s a tremendous cash burn?! Really?! Better trade on that arcane knowledge!). I think it’s important to answer these first two questions before getting into the specifics here. I’m going to try to remain both dispassionate and whatever is the opposite of myopic.
Debates I’m Not (Yet) Engaging In
I think it’s natural to start with a “forest-level” analysis before examining each specific “tree”, because it’s easy to get bogged down in thinking about a specific issue and then mired in a debate that may be irrelevant. This approach is probably going to attract criticism from both bulls and bears here, but I think it’s appropriate to start with a look at the available market size and the assumptions embedded in the stock. Getting immediately into the weeds distracts us from what I think are much more important issues.
Two “hot” debate topics are particularly irrelevant in my view, and I’m therefore going to actively ignore them. In honor of the industry under discussion, I’ve named each after an iconic TV commercial.
“Tastes Great, Less Filling,” or “The quality of Netflix content relative to its competitors.” I have very specific tastes, and I am often stupefied by what does well at the box office. I think Netflix has a host of great content that appeals to me and will keep me paying for the service. I also think some of the latest Disney (DIS) products are awful (I’m looking at you, various Star Wars reboots). That said, analysing businesses is about analysing the cash that they generate, and I can’t allow my own biases around quality to impact the analysis. I must give the Disney devil its due and acknowledge that its library is an impressive money maker. The reason I call this the “tastes great, less filling” debate relates to the fact that it’s possible for both companies to exist in the coming decades, given that the cost of each is insignificant to their potential markets. Thus, trying to pick apart the library of one or the other as the basis of an investment thesis is a wasted effort, in my view.
“You pay me now, or you pay me later,” or Netflix accounting and whether to capitalize or expense content creation. While I see both sides of this debate, I tend to favor capitalization. It’s not as though content like “Mind Hunter” will disappear at the end of an accounting period, so this is more asset than expense. Also, the interminable line up of Star Wars derivatives proves that these assets have staying power. I call this the “pay me now or pay me later” debate, because the assets get written down either immediately via expensing or over time via amortization. Over a long enough time line, each achieves the same purpose, and therefore, this debate is less relevant, in my view.
The Potential Size of The Market
The first thing I want to get a handle on is the potential size of the market, because this obviously puts an upper limit to growth. I think it’s important for investors to start thinking seriously about the global market here, because that’s where Netflix itself says its future lies. I exclude China, India, and Indonesia from the initial potential household count, because each country is enormous (combined, these three countries represent ~41% of global households) and deserve separate commentary. On the assumption that a single household will only purchase one subscription, I am very interested to know the number of households in each country. As I discovered, this is a more challenging thing to unpack than it would at first appear, for reasons that are obvious in hindsight. In the interest of full disclosure, I think the following are the weakest aspects of my demographic analysis.
A great deal of the data on household size is not current. For instance, the latest data we have on average Canadian household sizes comes from 2011. Population statistics are generally more current, but in order to work out the number of households in a country, we must divide the (more current) population data by the (more dated) average household size. I can accept this weakness, though, because household sizes vary much more slowly, and we may have reached lower limits of house sizes in the wealthier parts of the world. That said, I must acknowledge this inaccuracy in the global data.
There were 33 countries that I left out of the analysis for two reasons. First, they were too small to be relevant (sorry, Holy See). For example, fully 15 of these 33 countries have populations less than 1,000,000 people. Second, I omitted countries that are currently too poor to be considered for this analysis. The population of these 33 countries totaled just under 158 million people, or just under 2.3% of approximate current global population. Thus, I don’t consider the exclusion of these countries to be harmful to my estimates. That said, as the song says “the loser now will be later to win”, so at some point, these poor and populous countries will become relevant.
My analysis included just over 2 billion households worldwide, including China. Once I was able to establish the approximate number of households in the target countries, it then became necessary to organize these along income lines. Thankfully, I was able to find a source that listed each country along purchasing power parity lines. I divided the countries into those above and below median household income of $15,000. I awarded 70% of the households in each country with a median income above $15,000 to the “potential Netflix market.” In addition, I added 20% of the households of countries with median incomes between $7,000 and $15,000 to the potential Netflix market.
Based on my analysis of population, household sizes, and median income, I estimate that Netflix’s potential market, ex India, Indonesia and China, is somewhere around 355 million households +/- 20 million. The figure is obviously imprecise, but not to a massive degree, in my view. I’ve included the tables for this analysis at the end of this article, as they are large, and putting them here would break up the flow of the analysis.
The Special Cases of China, India and Indonesia
According to my data, China has ~454 million households, India has ~296 million and there are about 66 million households in Indonesia. Although each country has median income levels below my arbitrary cutoff of $7,000, each requires its own commentary. Although these three are enormous countries, I think they are far less relevant than their size would suggest.
A little over four years ago, investors heard that “Netflix is in talks to enter China.” A year later, we heard that the company was having trouble in China. It now seems unlikely that Netflix will be entering China anytime soon. In the background, iQiyi (IQ), the “Netflix of China” continues to gain market share. It seems that the Netflix strategy is to develop Mandarin programming (like “Wandering Earth”) to serve the millions of Chinese consumers living outside of China. For this reason, I’m excluding China from my analysis of potential households.
At first blush, India represents an enormous potential for Netflix to the Western observer, because we’ve all heard about the country’s rapidly growing “middle class.” In other words, it may be the case that my simple median income figures distort the full picture given the huge population that is living in abject poverty. The problem for the bullish case based on India is that the term “middle class” is largely irrelevant, as it includes diverse groups who are objectively poor but optimistic about their futures. According to Professor Anirudh Krishna of Duke University, if we settle the lower bound of what it means to be middle class in India to $10 per day, then only ~2% of the country’s population qualifies. That’s still 27 million people, which is significant but certainly not game changing. This suggests to me that at ~800 rupees per month ($11), Netflix will find a market, but it remains a bit rich for the vast majority of Indian families, in my view. That is part of the reason why a number of competitors have opened up specifically to serve this gargantuan market and why competition will likely remain fierce for some time. Given the alternatives available to Indian consumers, and given that they remain overwhelmingly poor, I’d add ~10 million potential Netflix subscribers in India. This is a reflection of the 27 million or so Indians living on $10 per day or more.
Indonesia is interesting because Netflix was blocked by the government-owned ISP back in 2016, though it’s still possible to access the service from other services. The government suggested that the move was because Netflix content has too much violence and “adult content.” It may also be related to the fact that the Indonesian government is (surprised gasp) slightly shady. In any event, the majority of Indonesians living on the nation’s 14,752 islands lack the bandwidth to stream the service, so I’m suggesting that for the foreseeable future, Indonesia doesn’t represent much of a potential market for Netflix services.
With the addition of 10 million potential subscribers in India, I am suggesting that Netflix has a total potential global market of ~365 million give or take 20 million. Given that the company currently has ~160 million subscribers, I would suggest that there’s another 205 million potential subscribers give or take 20 million households.
I think it almost goes without saying that picking a company that is growing sales rapidly is only part of what it takes to succeed at investing. It’s about much more than finding a company that’s growing cash flows at a decent clip and blindly buying that name. The price that we pay for that stream of future cash flows will obviously impact our long-term returns. For example, the people who paid $410 per share for their Netflix shares in July 2018 will have a much different long-term return than the people who paid $160 for the same shares exactly one year earlier. For that reason, I need to spend some time looking at the stock itself as a thing distinct from the overall business. When I look at the stock, I want specifically to spot the disconnect between expectations and reality. If the market for a given company is too optimistic, I want to avoid the name (or buy puts on it), because the shares will inevitably drop in price. The idea here is that if the market assumes perfection from the business, anything less than perfect execution will cause the shares to tank, as Netflix stock has done over the past few months.
In my view – and I hope this isn’t a controversial statement – a stock is worth the net present value of all future earnings that an investor can pull out of the company. When I look at stock prices, I don’t try to forecast the future, as that’s really hard and I’m really lazy. Rather, I try work out what rate of future growth is necessary to justify the current price. This allows me to work entirely with “known” variables (like stock price) and reverse-engineer the future growth that price must be assuming.
I can undertake this task in two ways, one more simple and one more complex.
The More Simple Approach
In terms of the simple approach, I’ll work out a growth rate in earnings that’s required to make future cash flows equal to Netflix’s current Enterprise Value and Market Capitalization. The current Enterprise Value is ~$162.3 billion, and the market capitalization is about $125.4 billion. Over the past four quarters, the company has earned $1.414 billion. Applying a 10% discount rate to future earnings, they would need to grow at a CAGR of about 28% for 20 years to match the current enterprise value, and they’d need to grow at a CAGR of “only” ~25.6% to match the current market capitalization. In my view, these growth rates are extraordinary and put a number on the level of optimism embedded in the stock price.
The following table is the calculation that matches Netflix’s earnings to enterprise value
(Source: Author’s calculations)
This is the calculation that matches the future earnings to market capitalization.
(Source: Author’s calculations)
The More Complex Approach
The other way in which I judge the assumptions embedded in current price is to turn to the methodology outlined by Professor Stephen Penman in his excellent book, Accounting for Value. In the book, Penman demonstrates how it’s possible to employ a fairly standard finance formula and (using the magic of high school algebra) isolate the “g” (growth) variable to work out what the market must be assuming about the company’s future growth trajectory. When I apply this methodology to Netflix, I’m not surprised by the result. The market is very optimistic about these shares going forward. In particular, applying this methodology to Netflix generates a similar conclusion to the simple approach. The market is very, very optimistic about the company’s future. Specifically, this approach suggests that the market is assuming perpetual (i.e., forever) growth of ~7.7% from Netflix.
For those who suggest that valuation doesn’t matter in this case given that the name has been “expensive” for some time, I would say that we’ve heard that story before and it’s always ended the same way. Netflix shares seem to be massively optimistically priced, and it should be noted that no company of Netflix’s current size has managed to grow earnings at either of these rates. These implied growth rates are even more challenging because, in my view, Netflix has “only” another 205 million potential customers on the planet. Just because no company has accomplished this feat, though, doesn’t mean it’s impossible. The company may also come up with other means to monetize its customer base going forward. In my next chapter on Netflix, I’ll dig into the company itself and address all of the issues found “in the weeds” here.
I conclude this first chapter of my Netflix analysis with what I think is a better understanding of the company’s potential market. There is a tremendous amount of further growth potential here, but it obviously has a limit. Matching the size of the upper limit to the forecasts baked into the stock price would seem to present a significant challenge to the long thesis. In my next look at Netflix, I’ll get into the weeds and see to what extent the optimism is warranted or not. For now, I remain neutral on the name. I eagerly await comments on how I might make this first analysis more robust. If your comments can help make me money or save me from a loss, I welcome them.
Appendix: Population Tables Used
Sources for household size can be found here. Sources for population can be found here. For some reason, household sizes for Denmark and Iceland were excluded from the first data set but can be found here and here. The following is the source for country median income.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: If, at the end of my three analyses on Netflix I’m either bullish or bearish, I’ll express that view with either calls or puts. Given that I’m trying to keep an open mind at this point, I have no position yet.