3 ms·

Buffett's whole approach to valuing companies to invest in doesn't really work for tech cos, especially software.

by tedmiston 9y ago

coldtea 9y ago
Prudence?
- tedmiston 9y ago
  More specifically, his approach of value investing / choosing to invest in companies per an intrinsic value based on fundamentals. https://www.investopedia.com/articles/01/071801.asp https://www.investopedia.com/articles/01/071801.asp
adventured 9y ago
It works on some fronts. He particularly likes to invest into businesses with moats. It's a theme he has spoken about constantly for decades. His personal problem is he doesn't understand technology businesses well enough to identify what those are or when they exist. It's equivalent to the specialized knowledge required to understand which biotech companies have competitive advantages or not, Buffett would also fail on that knowledge front (that is, he'd fail at it, not due to the lack of moats existing, but due to his own inability to understand the businesses). Microsoft, Intel, Cisco, Facebook, eBay, Amazon, Google - all tech companies with massive moats around them (now or in the recent past) that have helped to protect their core monopolies (or quasi-monopolies). Microsoft had a ~13 PE ratio five years ago (while paying a dividend), Apple was down to something like an 11 PE (pre the recent market bubble run); those are the kind of valuations Buffett likes in quality companies. So why'd he wait to invest into Apple until after it had already gone up so much (why didn't he buy it early into the iPhone ramp)? For the same reason he made a bad investment mistake with IBM, he doesn't understand the technology business at all (he thought Watson was some kind of special competitive advantage, that it'd be a dynamite business etc; he had no idea what he was talking about, it was out of his expertise to pick the winners in that area). Microsoft had an epic moat circa 1995. Buffett was friends with Gates at the time. Why didn't he invest? Microsoft was an adjusted $5 / share in 1995, it's $85 now (and has paid out a lot in dividends etc). In that time it has generated several hundred billion dollars in profit. Buffett has publicly stated it was because he had no particular ability to understand their business, to identify its competitive advantages; that doesn't mean those things didn't exist. Others with specialized knowledge of tech could identify those advantages. Google as another example has had an extraordinarily obvious monopoly moat in search for 13 or so years at this point.
sumedh 9y ago
Why doesn't' it work for software?
- nostrademons 9y ago
  He's stated it's because he doesn't understand it, but unpacking that... Buffett's investment strategy is to run a discounted cash-flow analysis on the stock's fundamentals to figure out its "true" value, and then only invest in it if this is significantly above its market price. Most consumer software companies have cash-flow graphs where they spend 5-10 years with zero revenue just gaining users and building a monopoly, then they turn on the monetization strategy and instantly start making $10B/quarter or so, and then they continue to make massive profits until some younger, cooler software company comes along and takes all their users, and then they either buy that company for a good fraction of outstanding shares or they die. In other words, their cash flows are unpredictable and back-loaded, and depend upon a number of factors that are unknowable early in the company's life, like 1) is there a viable monetization strategy? 2) when will the founder/CEO turn it on? 3) will they run out of capital first? 4) how big will the userbase be at the time? 5) will they be replaced by further developments in technology? For an investor who wants predictability and rationality in investing, these are anathema; it's like going to a casino where the house holds the information advantage and hoping you can play enough rounds that you win.
- tedmiston 9y ago
  The main argument for why value investing doesn't work in the tech sector is its basis on revenue trends over time. The whole approach is based on figuring out when a good time to buy a stock is based on when it's cheap / undervalued. However, many techs cos are growth stage investments, and it's hard to predict which ones will (eventually) win / succeed in creating moats. Determining the value of pre-revenue companies, for example recent tech IPOs, is just hard. You don't have the revenue history that you do for a company like Coca Cola. There's much more projection and extrapolation required. https://www.quora.com/Is-there-such-thing-as-a-strategy-like-Ben-Grahams-value-investing-but-in-the-technology-sector https://www.quora.com/Is-there-such-thing-as-a-strategy-like... https://www.mckinsey.com/business-functions/strategy-and-corporate-finance/our-insights/valuing-high-tech-companies https://www.mckinsey.com/business-functions/strategy-and-cor...

Buffett's whole approach to valuing companies to invest in doesn't really work for tech cos, especially software.

coldtea 9y ago

Prudence?

tedmiston 9y ago
More specifically, his approach of value investing / choosing to invest in companies per an intrinsic value based on fundamentals. https://www.investopedia.com/articles/01/071801.asp https://www.investopedia.com/articles/01/071801.asp

adventured 9y ago

It works on some fronts. He particularly likes to invest into businesses with moats. It's a theme he has spoken about constantly for decades. His personal problem is he doesn't understand technology businesses well enough to identify what those are or when they exist. It's equivalent to the specialized knowledge required to understand which biotech companies have competitive advantages or not, Buffett would also fail on that knowledge front (that is, he'd fail at it, not due to the lack of moats existing, but due to his own inability to understand the businesses). Microsoft, Intel, Cisco, Facebook, eBay, Amazon, Google - all tech companies with massive moats around them (now or in the recent past) that have helped to protect their core monopolies (or quasi-monopolies). Microsoft had a ~13 PE ratio five years ago (while paying a dividend), Apple was down to something like an 11 PE (pre the recent market bubble run); those are the kind of valuations Buffett likes in quality companies. So why'd he wait to invest into Apple until after it had already gone up so much (why didn't he buy it early into the iPhone ramp)? For the same reason he made a bad investment mistake with IBM, he doesn't understand the technology business at all (he thought Watson was some kind of special competitive advantage, that it'd be a dynamite business etc; he had no idea what he was talking about, it was out of his expertise to pick the winners in that area). Microsoft had an epic moat circa 1995. Buffett was friends with Gates at the time. Why didn't he invest? Microsoft was an adjusted $5 / share in 1995, it's $85 now (and has paid out a lot in dividends etc). In that time it has generated several hundred billion dollars in profit. Buffett has publicly stated it was because he had no particular ability to understand their business, to identify its competitive advantages; that doesn't mean those things didn't exist. Others with specialized knowledge of tech could identify those advantages. Google as another example has had an extraordinarily obvious monopoly moat in search for 13 or so years at this point.

sumedh 9y ago

Why doesn't' it work for software?

nostrademons 9y ago
He's stated it's because he doesn't understand it, but unpacking that... Buffett's investment strategy is to run a discounted cash-flow analysis on the stock's fundamentals to figure out its "true" value, and then only invest in it if this is significantly above its market price. Most consumer software companies have cash-flow graphs where they spend 5-10 years with zero revenue just gaining users and building a monopoly, then they turn on the monetization strategy and instantly start making $10B/quarter or so, and then they continue to make massive profits until some younger, cooler software company comes along and takes all their users, and then they either buy that company for a good fraction of outstanding shares or they die. In other words, their cash flows are unpredictable and back-loaded, and depend upon a number of factors that are unknowable early in the company's life, like 1) is there a viable monetization strategy? 2) when will the founder/CEO turn it on? 3) will they run out of capital first? 4) how big will the userbase be at the time? 5) will they be replaced by further developments in technology? For an investor who wants predictability and rationality in investing, these are anathema; it's like going to a casino where the house holds the information advantage and hoping you can play enough rounds that you win.
tedmiston 9y ago
The main argument for why value investing doesn't work in the tech sector is its basis on revenue trends over time. The whole approach is based on figuring out when a good time to buy a stock is based on when it's cheap / undervalued. However, many techs cos are growth stage investments, and it's hard to predict which ones will (eventually) win / succeed in creating moats. Determining the value of pre-revenue companies, for example recent tech IPOs, is just hard. You don't have the revenue history that you do for a company like Coca Cola. There's much more projection and extrapolation required. https://www.quora.com/Is-there-such-thing-as-a-strategy-like-Ben-Grahams-value-investing-but-in-the-technology-sector https://www.quora.com/Is-there-such-thing-as-a-strategy-like... https://www.mckinsey.com/business-functions/strategy-and-corporate-finance/our-insights/valuing-high-tech-companies https://www.mckinsey.com/business-functions/strategy-and-cor...