Selling for Valuation

In a post last week I mentioned using a "margin of safety" when selling (an idea given to me by another investor) and was asked to expand on the idea. I often end up selling too early when I think a stock is expensive. One thing I am already doing to correct this is to sell in three stages rather than all in one go. Some use technical analysis to time exits (ie don't sell while the price is going up) and the existence of price momentum alone justifies both of these latter two approaches. However, I think there is more to it than this and that a key part of the solution lies in using a "margin of safety" for selling.

I have long struggled with the question of why I keep getting it wrong when selling on valuation grounds. The evidence is clear that I am wrong. I used to think that the shares I sold were just overvalued as a consequence of too much money flowing into the market or because sentiment and technical traders cause prices to overshoot on the upside. Over time I have seen many of the "expensive" shares I sold (eg ALU) reach multiples of my sell price whilst their earnings continue to surge ahead. This has made me realise that my prior rationales are not sufficiently explaining what is going on even if there is an element of truth in them. They provided me with the comfort of thinking there is nothing wrong with my valuation process. I was tricking myself to protect my ego.

I now think that my mindset when selling was wrong. When buying, it is obvious that the risk of making a bad decision is that the price falls. If you believe that prices reflect values over time then the only way you can lose money over the longterm is if you pay more than a company is worth. There are many things that are impossible to know which determine a company's worth because they are yet to happen. Therefore, the best we can do is estimate a range of values. In order to protect ourselves from known (but unquantifiable) and unknown risks we buy with a "margin of safety", a price which is significantly less than our expected value of the company.

I have been mistakenly valuing stocks in the same way when selling. However, the risk when selling is selling too cheaply and whereas I need to think about downside risk when buying I should think about upside risk when selling. Under-receiving is deleterious to my wealth just like overpaying and applying a "margin of safety" when selling addresses this.

You might be thinking there is also a risk of not selling and there is, just as there is a risk of not buying. (Those who have invested in the US stock market in recent years and did not pick the FANG stocks will know this second risk well.) But these risks are only relevant where you can identify superior alternatives. In other words it only makes sense to sell a stock before the price satisfies a selling "margin of safety" in order to buy another stock which is sufficiently undervalued.


  1. So, if I understand your concept correctly, a MoS is subtracted from the IV when buying, and added when selling? If I have understood, how do you decide on the quantum of posiive MoS when selling?

    1. That's correct. How do you decide on negative MoS when buying? I would say it depends on the circumstances. As with most things in investing it is a mistake to look for precision because it is usually an illusion. Practically, all I'm really saying is that I will switch from a conservative to an aggressive valuation when selling.

    2. I agree you need to add an amount to your IV when selling, but only because of the effects of taxation - not because of the need to maintain an aggressive valuation. In my mind, the question when selling becomes "will I earn more if I hold all of x, or sell and invest 0.7x" (adjusting for whatever the tax rate is).

      The thing isn't instantly worth more because you own it, so valuation shouldn't behave that way in my mind.

    3. It is not about the business being worth more, it is recognising that the valuation is imprecise. It can be too low or too high. When I buy I recognise it could be too high so buy at a discount. When I sell I haven't been recognising that it could be too low. That is what I am correcting.

    4. I have a RFR with an equity premium I use as a MoS when buying, I vary that slightly depending on conviction. I really like the idea of a MoS when selling, I am just going to have to think a bit more about how I would calculate it.

      I know there are also times I override my MoS in a buying situation, and I think the same would be true selling. I believe its important to have a defined strategy with strong rules - but to be flexible enough to be able to operate outside them on occasion!

    5. Absolutely, flexibility and discipline - order and chaos - yin and yang. Peterson taught me the answer is treading the tightrope between them

  2. Or perhaps it depends on the economics of the business, the total addressable market, the long term growth hook. When Buffett was asked at what PE he would sell an overvalued business, he said there weren’t many truly great businesses and that it was almost always a mistake to sell. I think this was in the 1990s, Coke was on a high PE (in the 30s from memory) and was buying back it’s own stock. His comment on that was that he supported the repurchase of shares by companies that had a long pathway of growth ahead of them. More than two decades later Berkshire still owns those shares and it’s worked out quite nicely. Wisetech (and Altium) is on an insane PE but has a vast opportunity ahead and a dedicated founder. What makes more sense, selling now to take a short term profit or holding the stock for the next 20 years? I’m trusting that it’s the latter.


Post a Comment