Margin of Safety

How it works

Engineers designing a bridge don’t build it to hold exactly the maximum expected load. They build it to hold 3–4 times that load. The extra capacity is the margin of safety — the buffer that absorbs surprises, miscalculations, and worst-case scenarios.

Benjamin Graham brought this concept from engineering to investing: only buy a stock when its price is significantly below your estimate of its intrinsic value. The gap between price and value is your margin of safety. If your estimate is wrong by 20%, you’re still okay.

The principle generalises to any domain. Set deadlines earlier than you need them. Keep more savings than your budget model requires. Plan for fewer attendees than you expect. In every case, you’re trading a small amount of efficiency for a large amount of resilience.

Case study: How the Sydney Opera House’s budget exploded without margin

In 1957, Danish architect Jorn Utzon won the competition to design the Sydney Opera House with an estimated cost of A$7 million and a completion date of 1963. There was no margin of safety built into either estimate.

The revolutionary sail-shaped roof design proved far more difficult to engineer than anyone anticipated. Construction challenges, design changes, and political conflicts accumulated. The Opera House was finally completed in 1973 — ten years late — at a cost of A$102 million, more than fourteen times the original estimate.

Benjamin Graham’s concept of margin of safety — building a buffer between your estimate and your commitment — applies beyond investing. Had the original estimates included a 3x margin (A$21 million, completion by 1969), the project would still have overrun, but the political crisis and public fury over the cost blowout might have been avoided. In complex projects, the margin of safety isn’t pessimism. It’s realism about our inability to predict.

Real-world examples

Personal finance. If your monthly expenses are $3,000, a margin-of-safety approach means keeping 6–12 months of expenses in savings ($18,000–$36,000), not the bare minimum. It feels inefficient until the car breaks down the same month your company restructures.

Project timelines. If you estimate a project will take 8 weeks, tell stakeholders 10–12 weeks. The extra weeks absorb the unexpected problems that always appear. You’ll either deliver “early” (building trust) or on time (rather than late).

When to use it

Apply margin of safety whenever the cost of failure is high and the cost of the buffer is low, when you’re operating in uncertain environments, when your estimates are based on limited data, and when you can’t afford to be wrong (health, finances, relationships, career transitions).

Common mistakes

The main mistake is applying it uniformly. Not everything needs a margin of safety — low-stakes decisions with cheap reversibility don’t warrant the overhead. The second mistake is confusing margin of safety with pessimism. You’re not predicting failure. You’re designing a system that works even when predictions are wrong.

Try it now

Look at your current plan for something important — a financial goal, a project timeline, a career move. Ask: “Does this plan only work if everything goes right?” If yes, where can you build in a buffer? What would a 20% margin of safety look like?

Related models

These models complement Margin of Safety — they address similar situations from different angles.