• Don't miss out on our most popular articles saved in editor's pick. here

Beware the margin of danger


Beware the margin of danger

Most investors are familiar with the value investing mantra of buying a dollar for fifty cents. We call this discount of price over intrinsic value the margin of safety, because the bigger the discount, the more leeway investors have to account for the possibility that their estimate of intrinsic value is wrong. A wide margin of safety gives investors comfort. However, the margin of safety is only one side of the coin; on the other side is the margin of danger.


In a world where publicly available information is accessible at one’s fingertips and thousands of investment analysts are scouring the same investable universe looking for bargains, the margin of safety starts to give way to the margin of danger. In short, the margin of danger means that as markets become more efficient, the more an investor’s assessment of intrinsic value deviates from price, the higher the probability that the valuation is wrong. Put differently, a stock with a 50 per cent margin of safety isn’t immediately a bargain; it is first and foremost a signal that the investor has likely overestimated the value of the company, either through careless mistake, unjustified optimism, or simply misunderstanding the business model. To ignore this possibility is to arrogantly believe that one’s variant perceptions are always superior (i.e. different and right) to the market.

Fortunately, there are coping mechanisms available to help investors deal with the margin of danger. The global team uses the following mechanisms (amongst others) to minimize the probability and consequences of misestimating intrinsic value.

  • Use market implied expectations for growth and margins as a baseline for forecasts. Most investors will find it easier to gauge whether market implied expectations are reasonable, than to form their own set of sensible assumptions from scratch. This also focuses the investor’s attention on the assumptions underlying the intrinsic value, rather than the intrinsic value itself.
  • Take into consideration a stock’s valuation range as well as the level of conviction when sizing positions. The bigger the discount to intrinsic value (i.e. margin of safety), the lower the conviction in the valuation should be, unless there are compelling reasons for the investor to have a high level of conviction. A smaller position size minimizes the consequences of overestimating intrinsic value, and gives the stock more room to run before it becomes an oversized position.
  • Start each day with a blank mental portfolio, and ask yourself if you would buy back yesterday’s positions at the current market price given everything you know up to today (assuming no transaction costs and infinite liquidity). If the answer is “no” for any stock, that stock should be removed from the portfolio. This exercise helps investors to not only detach themselves from historical price movements, but also to regularly think about and update the margin of safety/margin of danger.

Daniel Wu is a Research Analyst at Montgomery Global Investment Management. Prior to joining Montgomery in June 2016, Daniel was an analyst in the investment banking divisions of UBS and Goldman Sachs, where he covered the Infrastructure, Utilities, Technology and Media sectors.


Why every investor should read Roger’s book VALUE.ABLE


find out more


This post was contributed by a representative of Montgomery Investment Management Pty Limited (AFSL No. 354564) and may contain general financial advice that is prepared without taking into account your personal objectives, financial circumstances or needs. Because of this, before acting on any of the information provided, you should always consider its appropriateness in light of your personal objectives, financial circumstances and needs and should consider seeking advice from a financial advisor if necessary.


  1. Hi Justin, I think that for a very good fund manager, a hard limit of 10% in a position (or in a bunch of closely correlated positions) is good risk management. Of course, for Buffett, Soros or Druckenmiller, this need not apply, but how many of us are Buffett/ Soros/ Druckenmiller? I can’t think of a single Australian fund manager that rivals their ability/ track record. I think they’re the exception rather than the rule.

  2. Thanks for your reply and explanation Daniel.

    Good to see you are meticulous and picked up on my mistake – I realized after I sent it that it was wrong – I meant to say ” the higher (greater) the margin of safety the better the outcome”


  3. Hi Daniel

    Your comments make sense and we all have a lot to learn from each other , but at the same time allowing MR Market to set the benchmark off which a margin of safety is based has it’s limitations. The comment I regularly hear from Roger is that the higher the price you pay the lower your future investment returns, so it SHOULD follow that the lower the margin of safety the better the outcome – PROVIDED of course MR Market has got it wrong. The reality is that MR Market is the product of the collective actions of thousands of buyers ans sellers who themselves are not always motivated by investment fundamentals , so relying on MR Market is not wise.

    A book that would be worth reading on the subject ( if you can get your hands on a copy) would be

    All the best for the festive season – I may not always agree with the views of the Team at Montgomery but value what I have learnt – we all learn from each other.


    • Thanks Max.

      The margin of safety can be loosely defined as the delta between the prevailing price of an asset and your estimate of said asset’s intrinsic value. Ergo, Mr Market will always set the “benchmark” off which the margin of safety is calculated. There are of course qualitative factors that contribute to the margin of safety, such as business quality, economic moat etc., but these should already be captured by your assessment of intrinsic value.

      Roger’s comment about higher price translating to lower future investment returns can be demonstrated as follows: assume we know with 100% certainty that a business has an intrinsic value of $100/share. If Mr Market is asking a price of $75, then the margin of safety is 25% (actually in this specific example margin of safety would be irrelevant because we know for certain the business is worth $100) and the future investment return is 33%. If Mr Market is asking a price of $50, then the margin of safety is 50%, and the future investment return is 100%. Therefore, it follows that the higher the margin of safety, the better the outcome assuming your assessment of intrinsic value is right and/or Mr Market is wrong (and not the other way around as your comment suggested). Roger’s comment can then be translated as “if you pay $75 for a business worth $100 (lower margin of safety), your future investment return will be lower than if you had paid $50 (higher margin of safety)”.

      Remember that the most important thing when it comes to investing is understanding the relationship between price and value. Investors have no choice but to rely on Mr Market for one of those two variables.

      I have read Margin of Safety and agree that it is one of the best books on the topic of value investing.

  4. Daniel, you write that “… a stock with a 50 per cent margin of safety isn’t immediately a bargain; it is first and foremost a signal that the investor has likely overestimated the value of the company…”.

    I think that stocks trading at a genuine 50% plus margin of safety are more common than you’re giving credit for – provided, of course, that you’re prepared to venture a little further afield than simply buying stocks in the mould of Buffett’s “wonderful companies”. Often, the margin of safety in companies exhibiting discounts of this degree is not particularly controversial. Most investors recognise that the stock is cheap but they overestimate or exaggerate the “cloud” then hanging over the company that has given rise to the discount.

    One of the best examples of this was Buffett’s 1973 purchase of stock in Washington Post. In his essay, “The Superinvestors of Graham-and-Doddsville”, Buffett said this:

    “The Washington Post Company in 1973 was selling for $80 million in the market. At the time, that day, you could have sold the assets to any one of ten buyers for not less than $400 million, probably appreciably more. The company owned the Post, Newsweek, plus several television stations in major markets…”

    Buffett estimated that just the value of the Post’s assets were $400 million at a time when it was selling in the market for $80 million. The value of assets are a lot easier to value accurately (and admit of much less controversy) than the value of future cash flows. It is also noteworthy that after his initial purchase the value of Buffett’s investment in the Post sank 25%.

    Now the cloud hanging over the Post at the time was, as Buffett later recalled, the fact that then U.S. President Richard Nixon had “sort of declared war” on the newspaper over its coverage of the break-in at Democratic National Committee offices in the Watergate complex. The Nixon administration orchestrated efforts to challenge the Post’s ownership of its lucrative Florida TV station and to incite shareholders to “go after” publisher Katharine Graham by targeting The Post’s real estate investments.

    These issues were capable of having very serious consequences for the value of the company. But their effect on the value of the Post was exaggerated. Most investors agreed that the Post’s assets alone were worth multiples of its then market price and yet they still held back from buying. Buffett, by contrast, did not merely take a nibble at a cheap stock but he loaded up – and that is perhaps an even more valuable investment lesson.

    When you see something that is a terrific value opportunity, you back up the truck. All that stuff about position-sizing makes no sense when the market is handing you value with no regard for its price.

    • Thanks for your comment Justin.

      Graham and Buffett invested in eras very different to today’s environment, so an example from 1973 may not be the most relevant. Information asymmetry is much lower today and markets are more efficient at pricing assets. If value investors stuck to investing in nets and net nets, they would have a much smaller investable universe today.

      Because prices are more efficient at incorporating available information (leaving aside the market’s tendency to under or overreact), investors should be cautious when they think they have found a “genuine” 50% margin of safety. Nothing makes that margin of safety “genuine” beyond the investor’s belief that they are right and all other investors/speculators are wrong. Investors should be even more cautious if they more often than not find a 50% margin of safety whenever they look at a stock. At Montgomery, we only invest in quality companies with strong prospects, so venturing further afield to buy poor quality cigar butts is not something we do. Anyone can make a case that the swing coal or oil producer is a 5-bagger if prices rebound to $100 – but that’s not a margin of safety, nor is it investing.

      As for your comment about position sizing making no sense, we invest our clients’ hard-earned money, so we are in the business of managing risk. An individual investor has much more leeway to back up the truck and take a 50% position on what they perceive to be a 50% margin of safety, and chalk it up to an expensive lesson learnt with no further consequences if the market turns out to be right.

      • Thanks Daniel. I would be very worried if the Montgomery Fund starts to take 50% positions in anything!

      • Daniel, you write that: “Graham and Buffett invested in eras very different to today’s environment, so an example from 1973 may not be the most environment”.

        I have two points to make about this.

        The first is that what I was highlighting through the example of Buffett’s purchase of the Washington Post Company in 1973 was the tendency of investors to overestimate or to exaggerate an issue that then clouds the outlook for a company. That tendency has not changed between 1973 and now, nor will it change. It is human nature to accord passing circumstances greater importance than they usually deserve.

        I will give you an example. Assured Guaranty Ltd (AGO) trades on the New York Stock Exchange. It is a bond insurer. It’s not a very sexy business. It was hit badly during the GFC but it survived and, through litigation, has managed to recover many millions of dollars in damages from banks and financial institutions which misrepresented to AGO the risk that it had insured up to 2007. Since then, AGO has continued to grow, slowly but steadily, both through acquisition and through a pick-up in the issuance of municipal bonds. It earns a very respectable return on equity. Since 2010, it has been around 15%. AGO is also competently managed. It does, however, have an exposure to various Puerto Rico government bonds. Puerto Rico has defaulted on some of those bonds. Other bonds it has sought to restructure. Consequently, AGO, pursuant to its policy, is required to pay the bondholders interest bi-annually on these bonds.

        But Puerto Rico accounts for 1.7% of AGO’s total municipal book. Pursuant to AGO’s policy, interest and principal repayments are not accelerated. Consequently, AGO’s exposure is spread out over the next 30 years. AGO’s Puerto Rico exposure, assuming it results in a complete loss and nil subrogated recoveries for AGO, is thus quite quantifiable and poses zero chance of any significant or long-lasting impairment to AGO’s business. Despite that, AGO’s share price has, until quite recently, traded at levels that implied that its Puerto Rico exposure was capable of wiping out half AGO’s business value: at the beginning of this year, AGO had tangible book value (conservatively estimated) of $44 which has been growing at 15% a year. It’s stock price was around $24.50 at the time. It’s P/E was 5.

        We can quibble about whether the margin at which AGO was trading in January 2016 to fair value was 50% or less. But it cannot seriously be disputed, I think, that in January 2016 AGO’s margin of safety was very wide and the chance of losing money by investing in AGO at that level was very low.

        This leads me to my second point. Saying that “nothing makes [a] margin of safety ‘genuine’ beyond the investor’s belief that they are right” is mistaken if it implies that asset values are purely matters of subjective opinion. If asset values are not purely matters of subjective opinion but can be determined reasonably accurately within a range, then it follows that the price for an asset and its deviation from the value of the asset can also be determined reasonably accurately. The deviation of price from the value of an asset is the condition that makes a margin of safety possible.

        By referring to a margin of safety as being “genuine”, I meant that the margin between price and value was readily apparent and did not admit of any real dispute. The example that I gave above of AGO trading at $24.50 in January this year when it had a tangible book value of $44 which had grown at 15% for the last 6 years is, in my view, an example of a genuine margin of safety.

        Your final point about Montgomery being “in the business of managing risk” so it is not able to thoroughly gorge on value opportunities when they arise is a curious inversion.

        Given the relative infrequency of value opportunities, might not another way to manage risk be to really go for an opportunity when it screams at you “value and safety” so as to put together year or two of really outstanding returns of 30%, 40% or 50%, so that when a lean year or two comes along you (and your investors) will have those fat years behind you that will allow you to maintain a superior record of investment performance?

        When you look at the historical returns today of Buffett, Soros, Druckenmiller and Greenblatt, you are looking at guys who made concentrated bets. And how it makes any sense to put money in your 20th best idea because you’ve reached the 5% allocation limit of your 1st best idea seems to me to be a recipe for mediocrity, whether one is managing one’s own account or the account of others.

      • I should add, Daniel, and I am sure you’d agree, that the very idea that one can identify a stock having a margin of safety with as precise a figure as 50% (or any other number) is itself specious. Such a claim is inconsistent with the very concept of a margin of safety and the purpose which it is designed to serve – which is: to act as a defence against the irreducible imprecision of business valuation

        So, on that ground alone, one should be fairly sceptical if an investor claims to have identified a stock with a 50% margin of safety (and not a basis point more or a basis point less). Rather, the most that can really be said is that the margin of safety is wide, very wide, wide enough, not very wide or non-existent.

Post your comments