Sunday, June 11, 2017

Notes on Wachenheim’s “Common Stocks and Common Sense”

Ed Wachenheim runs Greenhaven Associates, a $6 billion value fund. He piled up an enviable track record over the past 20+ years, regularly making 20%+ a year. And he does this by investing in large caps like IBM and Lowe’s. I’ve been resorting to nanocaps to in my quest for returns, so I wanted to read his book and see how he did it.

The book, "Common Stocks & Common Sense" has an excellent format. It is a series of case studies, each about a particular company Wachenheim invested in.

He would typically start with some historical backgrounds for the industry and the company in question. Then he shows how he develops a thesis, problems the company was having (to the extent there were any), and how the idea played out the next couple years. Throughout these passages Wachenheim would sprinkle in bite sized philosophy about how to use sell-side research, his view toward interviewing management…etc.

We need more books like this. This is a little like “Alchemy of Finance” where Soros documented his real time decision making. If anything, this book provides interesting snapshots of some industries from 1990's to now.

This book was helpful in confirming some of the ideas I have on how to approach research, on top of some great insights. I will first share my observations about Wachenheim’s style, then list out some general takeaways.

Observations of Wachenheim’s style

  • Buy large cap liquid securities. 15-25 stocks in the portfolio, try to be fully invested all the time. 
  • Targets 20% return. This means individual ideas need much higher upside (say 60%+), because there will be losers also.
  • Ignore broad market moves, focus on fundamentals.
  • Not a buy and hold forever type of guy. Willing to hold for 2-3 years and sell when the thesis is played out.
  • In general stay away from growth stocks with high multiples. But also try to stay away from severely distressed companies. 
  • Typical type of plays:
    • Turnarounds/ leadership changes: e.g. IBM, Interstate Bakeries
    • Cyclicals: e.g. U.S. Home Corp, Centex, Southwest, Lowes, Whirlpool
    • Good companies with temporary issues: Union Pacific, Lowes, Boeing, Goldman Sachs 

Miscellaneous Notes and Takeaways

1) Focus on the upside first. Creativity is key in target identification.

Before doing a big deep dive, do a simple model and project 2-5 years out to see if there's enough upside to justify further work. Try to be creative in generating thesis - what can make revenue and margins go up? Can management cut cost here and there? What happens if input cost goes down…etc.

A lot of people say “focus on the downside and the upside will take care of itself”. But at least in the idea generation/filtering stage, you have to demand the upside - else you’re just wasting your time. The focus on downside can come after an idea passes the initial filtering and you’re doing deep dives and running scenarios.

2) Know the long term history of the business and industry. This helps you appreciate the structural difficulties and what it takes to fix them. 

IBM was an example in the book. In the 1990’s the company’s business model of selling, leasing, and servicing mainframes became outdated due to emergence of the PC. That left IBM with a bloated cost structure which old management was not willing to address. The buying opportunity came when Lou Gerstner became the new CEO and showed “the courage to take tough steps”.

Knowing the long term history also helped Wachenheim invest in cyclical companies. He bought Southwest Airlines in 2012, understanding that the decade of 2000's were miserable for airlines and forced them to cut capacity.

3) True insight comes from understand the unit economics of the business well.

Knowing the economics of Union Pacific helped Wachenheim differentiate between a temporary problem and a structural one. In UNP's case, congestions led to train delays, which hurt margins because revenue is a function of volume, while much of the expenses like labor are a function of time. When congestions were fixed, UNP’s margins improved and stock went along with it.

Another example is when he talked about Southwest Airline's capacity utilization. 83.1% utilization might seem like there's some excess capacity. But that is the average. Since unpopular flights (bad hours, remote locations...etc) are going to have many empty seats, the more popular flights are probably operating at full capacity and having wait lists.

4) Historical average P/E of the market is a little less than 16x. When valuing companies Wachenheim would use the 16x as a benchmark and adjust up and down based on business quality.

This is a better approach than using comparable multiples, which may be the most over-rated valuation methodology of all. “Peer comps” is an easy way for unscrupulous bankers and analysts to justify over-valuation in bubbly markets (“hey Alibaba trades at 56x PE so let me value this other shitty e-commerce stock with 45x PE! Look I’m really conservative here!”)

5) Use sell side analysts mostly as a gauge of market consensus, and think “do I see an upside here that consensus is not pricing in?” 

That makes sense because sell side guys get their feedback from buy side community, so these analyst reports in some way reflects prevailing opinions.

6) In my humble opinion, Wachenheim can screen stocks more effectively. He seems to run a lot of screens based on low valuation ratios, which he himself admitted is rarely productive, since these “cheap” stocks are usually cheap for a reason. 

I think you’re better off screening with charts and basic technical analysis. For example if you're looking for out of favor stocks, don't screen for low P/Es, low EV/EBIT...etc. Rather, find some price charts that dropped like a rock more than a year ago, underwent heavy capitulation selling, and have since showed price stabilization along with low volume (a sign of investor disinterest). What I typically do is screen for some minimum quality (low debt, high ROIC...etc) then run them through charts.

No comments:

Post a Comment