Here is what I wrote:
“What I find most interesting about the FAANGs (and which seems comparable to the anointed stocks of prior eras) is the discussion rarely comes back to intrinsic value. For example, at what price is Netflix a sell? And if you owned the stock five years ago (when it was trading at $10 per share), what was your answer back then? I would love to see a write-up from an investor that answers these questions.”
Before discussing Netflix, I shoud zoom out a bit and talk about my investment philosophy. In short, I think it is important to examine decision-making over time and under different conditions (for example, if the stock was 30% higher or lower than it is today). I think it is important to remember that what is happening now should not be viewed in isolation; it is a point in time along a continuum. That way of thinking points to something I view as a fundamental truth of investing: the decisions we make today (or have made in the past or will make in the future) should be consistent over time.
The questions posed above are supposed to make the Netflix investor consider the following: What could possibly explain a nearly 20x increase in a stock over a few years? Framed another way, what fair value estimate would you have come up with for Netflix back in 2012 that could justify still owning the stock in 2017? There are a couple explanations that come to mind: (1) it was absurdly cheap a few years ago; (2) there has been a material change in the business; (3) it is absurdly expensive today.
Netflix may be the worstexample I could have picked to try and make this point. I believe the company is unique in that there is a reasonable basis for the second argument: the transition to the streaming subscription business fundamentally changed the company. You could make a strong argument that the user economics and sustainability of the company’s moat have materially improved as a result of this evolution. In combination with unabated growth, maybe there is a case to be made for a significant increase in the value of the business (by significant, I mean multiples of its value from even a few years ago).
More often than not, I think you will find the case for the second argument is pretty weak: most businesses do not fundamentally change in such a short period of time (or at least not the type of businesses I am drawn to). That leaves us with the other explanations (or some combination of them): either the stock was too cheap a few years ago or it is too expensive today.
This is where consistency matters. Personally, I find value in building a model and coming to a fair value estimate because it forces you to be intellectually consistent over time. Without these tools, I fear “thesis creep” and “fair value creep,” particularly in a bull market: as stock prices continue their unimpeded march higher, you might catch yourself doing whatever’s necessary to justify holding on for just a little longer (especially if you have already sold other positions and paid capital gains – only to watch those stocks keep moving higher).
While it may be painful to miss additional gains, the alternative is abandoning a value investing framework. As Howard Marks (Trades, Portfolio) noted in his most recent memo, holding stocks as they move past your sell points is either completely illogical or a sign of investor error and lack of discipline – the type of behavior that is typical in a bull market. That does not interest me.
I think a fair value estimate keeps me honest by (very roughly) quantifying my return expectations and giving me a basis for considering my alternatives (the opportunity costs). Let’s say the return estimate of a security is roughly 10% per annum for the next decade. If we check back in a year and the stock has doubled, our return expectation for the next nine years has fallen to approximately 3% per annum (all else equal). Does it make sense to keep holding this security?
Maybe something has changed. Maybe that is why the stock doubled over the past year. How does that change compare to what I assumed in my original analysis or thesis? Do the updates to the model materially impact the fair value estimate of the business – and as importantly, if I am forced to make a significant update to my model based on financial results over the past year, do I truly have a good grasp on the underlying drivers of intrinsic value for this business?
The point, as mentioned above, is intellectual consistency over time. I think it is meaningless to think of yourself as a long-term investor if your opinions and conclusions are fleeting (especially if they are fleeting in the face of negative price action alone). The conclusions supporting your investments should rest on a foundation that does not crumble at the first sign of trouble.
If the foundation is solid, we must have the conviction to act on both ends: to buy shares even as the stock price moves against us and to sell shares as the price exceeds our estimate of fair value. Doing the latter is more difficult in the current environment. If history is any indication, we'll eventually reach a point where those that had the conviction to sell when the math no longer made sense will be rewarded for their discipline.