How to Sell a Stock and Other Food for Thought from Tweedy, Browne

While the question of when to sell a stock is not something I’m particularly knowledgeable enough to answer, I’ve been doing some reading lately and discovered some thoughtful ideas about how to sell.  Tweedy Browne writes a lot of academic/empirical investing research that is very applicable and easy to understand.  Their paper What Has Worked In Investing is a classic in my opinion for corralling a collection of empiricism to the value investing process that is otherwise just intuitive and logical.  I recently read Investing For Higher After Tax Returns (h/t Whopper Investments) and while the main thrust is equally thought provoking and worthwhile, one of the corollaries struck me as having broader implications and as a worth while topic of further discussion.  While the paper focuses on the implication of investing for after-tax returns, one of the underlying themes is that how you sell a stock can have a large effect on your after tax returns and warrants serious consideration.

One reason cited by professional portfolio managers for selling is finding even more undervalued stocks.  An example would be having stocks A and B, which you think are both worth $10 and are for the purposes of this exercise identical businesses (margins, sales, ROE, etc).  You bought stock A for $6 and now it is trading for $8.  You don’t own stock B, but it is trading for $7.  Theoretically you should sell stock A and buy stock B as your potential return is 43% instead of 25%. 

The Tweedy paper would offer conditions to this, and I would agree.  Say stock A and B both earn $1/share and will grow 10% a year in perpetuity.  At their respective prices of $8 and $7 and a common value of $10 might be reason to sell.  If you sell stock A for $8, you receive $6.80 in after tax returns at the long term capital gains rate.  So you are selling a stock for proceeds that are 6.8x earnings, to buy a stock at 7x earnings.  This should remind you of Warren Buffett's criticism of the frozen pizza business sale by Kraft (Geoff's entire article is a great real world example of exactly what Tweedy is empirically pointing out)

You could truthfully say you bought stock A at $6/share and sold it for a $2 gain at $8/share and theoretically buy stock B at $7/share and sell it for a $3 gain if it goes to $10/share.  You could truthfully say you made $5 in gains.  Or could you?  Within the limits of this scenario, with the after tax proceeds you could not have even bought one share of stock B, since you only had $6.80 per in proceeds from stock A.  Clearly the closeness in valuation of stock A and B are a cause for pause and reflection.  In real life, no situation is guaranteed to be remotely close to this simplistic.  It might help to go to page 37-38 of the paper since the numbers are laid out in greater detail if you don’t find the above clear. 

While none of this will tell you exactly how to go about deciding when to sell, Tweedy (probably not me) does provide a cogent explanation that can at least serve as a starting point.  Depending on how you learn and think, their presentation through the use of a numbers in an example might help. 

I find that investors/bloggers/journalists that say one should sell a stock when it reaches one’s target price are not really telling you or I anything that is all that applicable.  Material events can happen that change the value of the stock for better or worse before your target price is ever reached.  What if you reassess the value and its higher, but at the current price you aren’t going to add more.  Blindly selling at an arbitrary price is not very scientific or thoughtful.  Selling when it reaches a price target oversimplifies something a lot more complex.  While Tweedy doesn’t exactly answer the question (they didn’t set out to and their paper is more about finding stocks that you won't be selling any time soon), they do provide a useful starting point to develop an understandable framework for how to sell your stocks, assuming you are going to buy another in its place.  


Some other interesting nuggets in the paper:
- A defense on their diversification, which I find compelling starting on page 43
- Checklist on earnings outlook in Appendix C starting on page 63
- Checklist on assessing growth prospect, competitive position, and economics of a business starting on page 65