Dealing with the Margin of Danger (Part II)

How can investors deal with the margin of danger?

The important thing for investors to accept is that the margin of danger comes hand in hand with the margin of safety. This means that we can never completely remove the margin of danger, but can only minimize the probability of the intrinsic value estimate being wrong, and mitigate the impact if it is wrong. That is, we want to flatten the curve from the dotted line to the solid blue line:

1

The Montaka team has used the below mechanisms (among others) since inception to help us manage the margin of danger by minimizing both the probability and consequences of our intrinsic value estimates being wrong.

  • Start with market implied expectations. When assessing a new idea, the first thing an investor should do is establish what market implied expectations for revenue growth and margins are baked into the current share price. Investors have a much easier time determining whether a set of given assumptions are unreasonable, than they do coming up with a set of sensible assumptions from scratch. This exercise helps mitigate anchoring bias by focusing the investor’s attention on the assumptions, rather than on whether the intrinsic value estimate “feels” reasonable and then playing with assumptions until it does. Furthermore, having a set of market implied expectations as a baseline forces the investor to articulate why his base case assumptions should deviate from market expectations.
  • When it comes to sizing long positions, Montaka employs an algorithm that takes into account a stock’s valuation range as well as our level of conviction in the valuation. Unless there is compelling evidence to support a high level of conviction (e.g. we have followed the company for a long time or the business model is straightforward), a stock that has a very wide margin of safety/danger should be sized at a lower conviction level. The smaller position sizing has two benefits: i) it minimizes the negative consequences of being wrong; and ii) it gives the share price runway to appreciate before the position reaches size limits and we need to cut for risk management purposes. Chris has written extensively about how we size portfolio positions, and the articles can be found here: Part 1, Part 2 and Part 3.
  • Finally, and perhaps most importantly, we have a unique mental framework that we use when reviewing the Montaka portfolios. We monitor and review our portfolios on a daily basis, but we do so by assuming we are building a new portfolio each day. That is to say, every day we assume Mr Market buys all of our positions at the prevailing prices, and then offers to sell them back to us at the same price (assuming no transaction costs). Based on all the information we know about each stock up to today, we then decide if we would buy them back.

This exercise helps investors detach themselves from historical price movements and the price they historically paid for the shares, because the only considerations relevant to investors are the current share price and their estimate of intrinsic value based on all information they know up to that point. This also prompts investors to constantly update the margins of safety on their stocks (and thus think about the margin of danger). If the decision is not to buy back a stock in this theoretical exercise, then perhaps the perceived margin of safety is wrong and the stock should be taken out of the portfolio. (We also repeat the same exercise for our short portfolio.)

Seth Klarman summarized the duality of value investing succinctly when he said that investors need to be arrogant enough to think the market is wrong and bet against it, but also humble enough to recognize when he is wrong and the market is right. Investing on the margin of safety is the arrogance; respecting the margin of danger is the humility.

DH5_2155Daniel Wu is a Research Analyst with Montgomery Global Investment Management. To learn more about Montaka, please call +612 7202 0100.

Leave a Comment

Your email address will not be published. Required fields are marked *