--- title: "Margin of Safety: What It Is, How to Size It, and Who It Protects" description: "A margin of safety is the gap between your estimate of value and the price you pay. How the formula works, how wide it should be, and why a market-wide discount is not your cushion." author: "Sarah Fontaine" category: "Investment Metrics" date: 2026-10-25T07:05:01.098Z canonical: "https://investormuse.com/blog/margin-of-safety-explained-price-to-fair-value" --- # Margin of Safety: What It Is, How to Size It, and Who It Protects ![market, produce, farmer's market](https://cdn.pixabay.com/photo/2022/03/27/10/18/market-7094635_1280.jpg?w=1200&q=75) > A margin of safety is the gap between your estimate of value and the price you pay. How the formula works, how wide it should be, and why a market-wide discount is not your cushion. Between the last day of 2025 and the third week of March 2026, the cushion on the US stock market roughly tripled. Morningstar's composite of the roughly 700-plus US-listed companies its analysts cover went from a price/fair value reading of 0.96, a 4% discount, to 0.88, a 12% discount ([Morningstar, Q1 2026 North America Market Outlook, data as of 31 December 2025](https://www.morningstar.com/content/cs-assets/v3/assets/blt9415ea4cc4157833/blt874a8929284ea035/6966a0879e0faf0008a3c71e/Q12026_NA_Market_Outlook.pdf); [Morningstar, Q2 2026 outlook, data as of 23 March 2026](https://www.morningstar.com.au/stocks/chart-week-us-equity-market-trading-an-attractive-discount)). Almost nothing about the companies changed in between. The margin of safety lives in the price, never in the business, and that sentence is the entire concept. ## What the Cushion Is Actually Absorbing Every valuation of a share misses by some distance, because it rests on future sales, future costs and a future mood nobody has seen yet. The cushion for that miss is what this idea names, and price is the only way to buy any. The arithmetic runs to one line. Take your estimate of value, subtract the price, and divide the result by the estimate. An estimate of $60 a share against a price of $45 leaves $15 of room, a quarter of the estimate. Pay $57 for the same share and the room is $3. Nothing about the company moved. Your tolerance for being wrong collapsed. Benjamin Graham, who put the phrase into general circulation, called that difference the safety margin and said it was there for ["adsorbing the effect of miscalculation or worse than average luck"](https://novelinvestor.com/grahams-margin-safety/). His gloss on it matters more than the formula: one security carries a large margin at one price, a thin one further up, and nothing at all above that. Follow the logic and a conclusion arrives that most stock commentary ignores. A fine business does not possess a margin of safety. A fine business at a stated price does. Warren Buffett set out the same rule in Berkshire Hathaway's 1992 letter to shareholders, writing that the company insists on a margin of safety in its purchase price and loses interest when its own calculation lands ["only slightly higher than its price"](https://www.berkshirehathaway.com/letters/1992.html). He credited the principle to Graham and called it ["the cornerstone of investment success"](https://www.berkshirehathaway.com/letters/1992.html). ## A Wide Gap Buys You Three Specific Mistakes - **A growth forecast that was too kind.** Sales come in slower than the model assumed, and the value estimate drops with them. - **A multiple that shrinks underneath you.** Borrowing costs rise or enthusiasm cools, and the market simply pays less for the same earnings. - **An event the model never held.** A lawsuit, a recall, a product that fails, a customer that leaves. What a wide gap will not do is set a date. Shares bought below value can sit there a long time. The cushion is protection against a permanent loss, not a timetable for a gain, and Graham was explicit that the margin gives a better chance of profit than of loss without making loss impossible, which is why he paired it with holding more than one position. ## A Narrow Gap Demands a Near-Exact Valuation On a thin cushion the valuation has to be close to exact, because a modest overestimate of value turns a reasonable purchase into a loss. Plenty of admired companies sit in that position, not because the businesses are weak but because today's price already pays in advance for the performance. Reading somebody's published fair value figure well enough to lean on it is a separate skill, and we take it apart in [what actually sits inside a fair value estimate](/how-to-read-a-fair-value-estimate). ## The Same Market, Two Quarters Apart Morningstar's quarterly snapshots make a verifiable example, because the firm publishes both the reading and the date it was taken. As of 31 December 2025 that coverage composite, upwards of 700 names listed in the United States, stood at a price/fair value of 0.96, which the firm described as a 4% discount ([Morningstar, Q1 2026 North America Market Outlook](https://www.morningstar.com/content/cs-assets/v3/assets/blt9415ea4cc4157833/blt874a8929284ea035/6966a0879e0faf0008a3c71e/Q12026_NA_Market_Outlook.pdf)). By 23 March 2026 the same composite read 0.88, a 12% discount, with large companies at 13%, mid-sized ones at 6% and small ones at 17% ([Morningstar, Q2 2026 outlook](https://www.morningstar.com.au/stocks/chart-week-us-equity-market-trading-an-attractive-discount)). Across those twelve weeks the firm did not decide that American business was worth less. Prices came down. The cushion widened because the numerator moved, which is the mechanical version of Graham's point. The direction can run the other way too: Morningstar had raised its own valuations across coverage by roughly $7 trillion during the fourth quarter of 2025, about 8% of the market capitalisation it covers, including increases of $2.5 trillion for Alphabet, $1.2 trillion for Nvidia, $740 billion for Broadcom and $440 billion for Apple ([Morningstar, Q1 2026 outlook](https://www.morningstar.com/content/cs-assets/v3/assets/blt9415ea4cc4157833/blt874a8929284ea035/6966a0879e0faf0008a3c71e/Q12026_NA_Market_Outlook.pdf)). A discount can open up because prices fell or because an analyst's estimate of value rose, and those two are not the same event. ## Three Stocks Were Holding the Whole Market's Discount The part worth arguing with comes out of that same report. Morningstar noted that lifting Nvidia out of the market calculation would move the price/fair value reading to 0.98, and lifting Alphabet and Broadcom out as well would put it at 1.00, exactly level with the composite of fair values ([Morningstar, Q1 2026 outlook](https://www.morningstar.com/content/cs-assets/v3/assets/blt9415ea4cc4157833/blt874a8929284ea035/6966a0879e0faf0008a3c71e/Q12026_NA_Market_Outlook.pdf)). The market-wide cushion at the end of 2025 was, in practical terms, three positions deep. The position taken here is that a market-level margin of safety is nearly useless as a cushion for a particular portfolio, and that quoting one is the most common misuse of Graham's idea. A cushion protects you against your own estimate being wrong about the things you hold. When the aggregate discount lives in three very large companies and you own none of them, you have the headline and none of the protection. The aggregate figure also invites a conclusion Graham's framing rules out, because it describes an index, and an index has no purchase price with your name on it. Hold a broad market fund and the aggregate figure genuinely is your figure, because the composite is roughly the thing you own. Even then the concentration deserves a second look, because a cushion carried by three companies is a bet on three companies dressed as a diversified discount, which is the failure we pull apart in [owning the same risk twice without noticing](/portfolio-diversification-mistakes-same-bet-twice). ## Four Ways the Idea Gets Misapplied Borrowing another investor's valuation comes first. A published fair value is the output of a model, and it carries that model's assumptions about growth, margins and the cost of capital. Studying one is useful; copying one is not, and two careful analysts looking at a single company routinely land on different answers. Second comes reading cheap as safe. A price that looks low against last year's profit sometimes marks a bargain and sometimes marks profit that is about to shrink, with the ratio flattering a business on the way down, a trap we set out in [why a P/E ratio misleads more often than it informs](/why-pe-lies-more-often-than-it-tells-truth). Third, skipping past how good the valuation itself is. A huge gap measured against careless work is worth nothing at all. Before you take comfort from the distance, ask whether the figure on the far side rests on reported results or on a story about them. Fourth, expecting the cushion to size itself. A company selling the same staple product year after year can carry a narrower gap than a fast-growing software business whose next three years could land almost anywhere, because the thing being measured is your own doubt. And a gap is only a cushion when the value behind it is real and reachable, which is the lesson of [a fund that trades below the assets it reports](/closed-end-fund-discount-to-nav-explained). ## The Price Is the Only Input You Control You cannot make a business more certain, and you cannot make a forecast sharper than the information behind it allows. You can decide what to pay. This idea is just the name for that decision, scaled to the size of your own doubt, and it is the one input nobody else sets for you. *This article is for educational and informational purposes only. It is not financial advice, and Investor Muse is not a registered investment advisor. Historical figures describe the past and do not predict future results. Speak with a qualified professional before you make decisions about your money.* *Position disclosure: neither Investor Muse nor its editorial team holds a position in any company, fund or security named in this article.* ## FAQ ### What is a margin of safety in investing? It is the distance between your estimate of what something is worth and the price you pay for it, usually written as a share of the estimate. An estimate of $60 against a price of $45 leaves a quarter of the estimate as room for error. ### How big should a margin of safety be? Wide enough to cover how wrong your estimate could be. A business with steady, repeat demand can carry a narrower gap than one whose growth could plausibly land anywhere in a broad range, because the second estimate has far more room to miss. ### What does price-to-fair-value mean? It is a share price divided by an analyst's estimate of fair value. A reading under 1 means the price sits below that estimate. Morningstar reported its US coverage composite at 0.96 as of 31 December 2025 and 0.88 as of 23 March 2026, which it described as discounts of 4% and 12%. ### Does a margin of safety mean I cannot lose money? No. Graham's own framing was that the margin gives a better chance of profit than of loss, not that loss becomes impossible. A position can keep falling after you buy it, and a wide gap measured against a careless valuation protects nothing. --- Source: https://investormuse.com/blog/margin-of-safety-explained-price-to-fair-value