Sunday, October 21, 2018

Value Investing 2.0 and 3.0

I consider myself a value investor, I have read several of the classic books on the topic and practiced it.

The original value investing, what Ben Graham practiced and thought, and what Warren Buffett started with I would now call Value Investing 1.0. Net-nets, low PB, low PE with nominal growth, and similar approaches worked then and can sometimes still work now. I think the risk currently is that the pace of change and ease of disrupting incumbents increased so much, that revenues and margins can decrease so quickly that such 'value' investment indeed becomes a value trap.
IBM, KHC (lately), retailers, etc. come to mind.
There will be of course good outcomes in that category, and I hope that BHC and VER will turn out to be like that in my portfolio.

Famously, Buffett thanks to Munger realized the value of quality/moat/growth and enhanced value investing to recognize these qualities. The prime example given by them is of course See's Candies.

The modern version of it are all the technology companies, which can grow revenues at a high rate with virtually no incremental capital required or when at worst R&D is the new Capex. Let's call it Value Investing 2.0.

Present-day FB and GOOG are ideal examples of that which I own. But many other companies fell into that bucket, with the increased importance of internet.


Lately I came across Whitney Tilson's interesting passage in his GOOG article:
Rather, it’s in part because the stock has always appeared expensive me, using the traditional valuation metrics with which I’m most comfortable • But there are other reasons, rooted more in emotion than logic: – As a contrarian and value investor, I don’t like owning what everyone else owns – I don’t like buying stocks that have already risen a lot (instead, I prefer to bottom-fish among the beaten-down stocks of out-of-favor companies, betting that they can turn things around) – I felt extreme regret for not having long ago purchased the stock of this incredible company – a classic case of the “I missed it” phenomenon

Buffett and Munger were asked, “…what have you learned about investing in technology companies?” Munger answered that their “worst mistake in the tech field” was not investing in Alphabet: • Well, we avoided the tech stocks, but as we felt we had no advantage there and other people did and I think that's a good idea not to play where the other people are better, but you know, if you ask me in retrospect, what was our worst mistake in the tech field, I think we were smart enough to figure out Google. Those ads worked so much better in the early days than anything else. So I would say that we failed you there and we weren't smart enough to do it and didn't do it. We do that all the time too.

On top of that, I listened to a great interview with Bill Nygren, which covered Netflix and how they would have only 14x PE in late 2017, should they increase price from $10 to $15. Another interesting angle on that is the approach to value users that Aswath Damodoran described here. In some way, this is of course what VCs always look for and try to value, at least during later funding rounds.

When done from a conservative perspective and based on proven track record and legitimate value that given service/product offers to customers with minimal marginal costs, I think this deserves a to be included in the 'value' world - as a distinct bucket, let's call it Value Investing 3.0. Post-IPO GOOG, FB and event recent NFLX per Bill's insight are I think prime examples of this.


This leaves my portfolio with one notable position which I did not bucket just yet - GM. That's because I believe it actually has pieces of each three. Let me explain:

1.0 - even though it's a cyclical stock, the current PE (adjusted for one-off charges) gives lots of safety and GM should break even at SAAR of 10-11M. With current SAAR in ~17M and average fleet age of over 11 years, I have a hard time imagining that we can reach 10M even in severe recession. The only reason it could happen is an ultra rapid adoption of self-driving across all types of vehicles...

2.0 - GM ain't internet business, but their capital allocation, discipline in building for the actual demand and keeping incentives at bay, as well as efficient capex and high ATPs make them a very strong player.

3.0 - I'm a big fan of self-driving and I think GM is best positioned to rapidly deploy in this space. Waymo is more advanced, but I think Cruise is catching up quickly and much better positioned to deploy quickly at scale than Waymo is. Integrating self-driving tech into an arms-length manufacturer's cars will be neither scalable nor cost effective. Investments from Vision Fund and Honda prove that this is a sound strategy.


AMZN is another outlier, which I actually think of as a combination of 2.0 and 3.0. While capital-intensive, the competitive moat of AMZN businesses is akin to Walmart decades ago - low-cost,  operator with huge scale that is almost impossible to fight with. But there is also a unique ability of Amazon/Bezos to expand into new areas and grab new markets. Such skill/ability is a truly unique thing.

Saturday, October 20, 2018

Shorting

I just listened to Mark Spiegel vs. Whitney Tilson discussion over short selling and I'm intrigued by the fact that Whitney decided/was forced to cover after the short trade went against him.

If the 'buy low, sell high' is the (value) investing way to make money, it would be natural to 'sell high, buy low' to be the way to run a short selling operation. Unless investment thesis changes (which it didn't in this case), it seems foolish to realize loss instead of waiting or doubling down.

This implies that short positions should be small so they can be allowed to grow if the trade (initially) goes against short seller.


Another curious aspect is that clearly many short sellers (with allegedly value investing approach) increase they short positions as the stock goes down. For me this is an equivalent of momentum/technical investing - directly against the value approach. I acknowledge that it is somehow different, when a short thesis catalyst appears that drives price down, but if the stock bounces back (as Tesla did numerous times) one can end up with 'sell low, buy high' result.


I just reviewed my trading history with Tesla short:
- I initially shorted by long dated puts and they either expired worthless or a remaining few are 50% off mark-to-market
- the later approach, of scaling the straight short position as TSLA goes up and reducing the short as it goes down was much more successful (and done on a bigger scale)

I'm torn whether to go back to the first approach as everything seems to indicate that the end is near for TSLA. On one hand lots (pun intended) of unsold inventory, Oct 15th '2018 tax-credit guarantee' and now introduction of mid-range M3 indicate quickly drying demand.
As a side note - my suspicion is that mid-range RWD M3 is actually programmatically-limited long-range RWD M3, partially taken from those parking lots to generate cash even at poor or non-existent margins.

My other short - SPY - was sold on avg. at ~290 and covered on avg. at ~286, with a very small position. Therefore it generated only a token profit, but still provided a reassurance during market swings and reduced anxiety created by using margin.
But during a more severe market decline this would hit me with even more rapidly increased long exposure than with just my basic approach of buying the dip and averaging down. And in exchange it is not likely to produce a meaningful profit. So for the time being I will forfeit this and focus on being long. TSLA is the only exception - the entertainment value alone is worth it.