You’ve learned how to find good companies at a discount. Now let’s talk about what happens when that discount disappears — and what to do about it.
We’ve spent eight posts building one thing: a way to answer the question what is this business actually worth?
You learned how to measure the cash it generates. How to account for time and risk. How to project five years of growth, add a terminal value for everything beyond that, and arrive at a single number — intrinsic value per share. And in Part 7, we did it for a real (fictional) business, Maple Ridge Manufacturing, and found an intrinsic value of $61.81 at a time when the stock was selling for $42. A 32% margin of safety.
That was the buy case. This post is the other half.
Every discipline that teaches you when to buy must also teach you when to sell. Without a clear sell framework, the margin of safety becomes a one-way door. You wait for the discount and walk in. But if you never have a reason to walk out, you’re not investing with a system — you’re hoping.
This post covers four distinct reasons to sell: when price exceeds intrinsic value by enough to invert the margin of safety, when the business itself becomes a wealth destroyer, when the original investment thesis falls apart, and when dividend reinvestment starts compounding the problem instead of the returns.
The 120% Boundary — When the Margin of Safety Inverts
Start with what the margin of safety is doing.
When we bought Maple Ridge at $42 against an intrinsic value of $61.81, the gap — $19.81, or about 32% — was our buffer. It said: even if my estimates are off by a meaningful amount, I’m probably still not overpaying. The margin of safety isn’t confidence. It’s honest doubt, priced in.
Now imagine Maple Ridge has climbed. The market re-rated it. Other investors figured out what you did, and then some. The stock is now at $74.
At $74, the math has changed fundamentally. Let’s redo it:
Intrinsic Value: $61.81
Current Price: $74.00
Premium over IV: $74 − $61.81 = $12.19
Premium %: $12.19 ÷ $61.81 = 20%
The stock is now trading at 120% of intrinsic value. The margin of safety isn’t just gone — it has inverted. You’re no longer getting a discount. You’re paying a premium.
Here’s the logic: the whole point of a margin of safety is that your estimate might be wrong. If the true IV turns out to be lower than your model says, you wanted a buffer to protect you. At $42 with IV of $61.81, that buffer was real. At $74, it’s reversed. You’re now hoping your IV estimate is too low — that the true value is higher than $61.81 — because that’s the only way buying at $74 makes sense.
That’s speculation. You’ve moved from “I’m buying a dollar for 68 cents” to “I’m buying a dollar and hoping it turns out to be worth $1.20.”
The 120% sell boundary is a practical rule: when the market price exceeds intrinsic value by 20% or more, the margin of safety has reversed, and the rational move is to exit. Not because the company is bad. Because the discipline that told you to buy at a discount is the same discipline that tells you to sell at a premium.
Think of it this way. If you found a house worth $400,000 and bought it for $270,000, you were disciplined. But if you then tried to sell it for $480,000 — and priced it as if it were still your great deal — you’d be deceiving yourself. The deal was the price, not the house. When the price is gone, the deal is gone.
The Wealth Destroyer — A Different Kind of Sell Signal
The 120% boundary is about price versus value. This signal is about what the business is doing with its own money.
Here’s the concept. Every business generates earnings. Then it faces a decision: pay those earnings to shareholders (as dividends or buybacks), or reinvest them in the business to create future growth. Most businesses do both — some goes out, some stays in.
The question that matters for you, as an owner, is: when the company reinvests a dollar, does it create more than a dollar of value?
This is measured by comparing two numbers:
– ROIC — Return on Invested Capital. The return the business earns on the money it reinvests. Think of it as the “yield” on reinvestment. If a company keeps $10 of your capital and earns $1.50 from it, the ROIC is 15%.
– WACC — Weighted Average Cost of Capital, which we built in Part 6. This is the minimum return shareholders (and lenders) require to stay invested. It’s the hurdle rate.
If ROIC > WACC, every dollar reinvested creates more than a dollar of value. The business is compounding wealth. This is good.
If ROIC < WACC, every dollar reinvested destroys value — even if the company reports positive earnings, even if the headlines look fine. The business is growing in size while shrinking in quality. This is a wealth destroyer, and it’s a sell signal independent of price.
Here’s why that distinction matters. Imagine Maple Ridge’s ROIC has been running at 12% while its WACC is 7.8%. Every dollar reinvested creates $1.12 in value. The business is compounding for you.
Now imagine a few things change over time. Competition intensifies. Pricing power erodes. The cost of inputs rises. Three years later, Maple Ridge’s ROIC has fallen to 6.5% — below its 7.8% WACC. Every dollar reinvested now generates only $0.65 of value per dollar of cost. Management is busy growing the business. But growing a wealth-destroying machine makes it bigger, not better.
And here’s the connection to your DCF model: ROIC below WACC doesn’t just signal a bad quarter — it changes the intrinsic value itself. The IV you calculated assumed a certain growth rate and a certain terminal value. Both of those assumptions implicitly depended on the company’s reinvestment being productive. When ROIC < WACC, your IV estimate is too high. The real value of the business, correctly modeled, is lower than what you calculated — and it declines further with every dollar the company reinvests at below-cost returns.
This means you can hold a stock at “fair value” while the business is actively deteriorating beneath you. That’s not neutral. That’s paying a full price for a shrinking asset.
A sustained ROIC < WACC — confirmed over multiple quarters, not a single bad report — is a reason to exit regardless of where the price sits relative to your IV.
When the Thesis Breaks
The DCF model you built is based on assumptions. Revenue growing at a certain rate. Margins holding at a certain floor. The competitive position remaining roughly intact. Those aren’t facts — they’re your best estimates, grounded in the evidence available when you bought.
When those assumptions prove wrong, you have to decide: is this a temporary blip, or has the story I bought actually changed?
There are two very different sell situations here, and conflating them leads to bad decisions:
Situation 1: The thesis is intact, but the stock has gotten expensive. This is the 120% case. The business is performing as expected. The market has noticed. The opportunity has closed. Sell because the price discipline says so — not because you’ve lost faith in the company.
Situation 2: The thesis has broken. This is more serious, and often more urgent. This is when the business results have consistently contradicted the assumptions you made when you bought.
What does a broken thesis look like? A few examples:
-
You modeled Maple Ridge’s free cash flow margin at 15%. For the past six quarters, they’ve delivered 8%. That’s not a blip. That’s the model being wrong. If 8% is the real baseline, the IV you calculated — using 15% — is overstated. Continuing to hold “because it’s still below my original IV” ignores that your original IV was built on a premise the business has now disproven.
-
You bought because the company had a durable competitive advantage — a brand, a patent, switching costs, some moat that protected pricing power. Then a competitor undercuts them on price. Then another. Their revenue holds for a year, then starts declining. The moat you valued is eroding. The terminal value in your model assumed that moat would persist. It won’t.
-
Management changes. A new CEO reverses the capital allocation strategy you respected. The reinvestment that was generating 14% ROIC gets redirected into acquisitions that immediately dilute returns.
In each of these cases, the sell trigger isn’t “price reached 120% of IV.” It’s “my IV estimate is now unreliable because the premise behind it has changed.” When the thesis breaks, holding is not neutral — it’s owning something that no longer matches the investment case you underwrote.
The key discipline: revisit the thesis periodically, not just the price. Earnings beats and misses happen. What matters is whether the structural case — the free cash flow trajectory, the competitive position, the quality of reinvestment — is intact or not.
The DRIP Trap
This one is subtle. It affects investors with long-hold, dividend-paying positions. And it matters more than most people realize.
DRIP stands for Dividend Reinvestment Plan — a feature most brokerages offer where your dividends are automatically used to buy additional shares of the same stock instead of landing as cash in your account.
DRIP is powerful when a stock is undervalued. Every reinvested dividend buys more shares at a discount. The compounding works for you.
But when the stock is overvalued — when it’s trading above your intrinsic value — DRIP compounds the error. Every reinvested dividend buys more shares at a premium. You’re making an active buying decision at the wrong price, automatically, every quarter.
Here’s the math. Say Maple Ridge pays a $1.50 annual dividend on 100 shares — $150/year. At $42 per share (32% below IV of $61.81), each reinvested dividend buys shares at a 32% discount. That’s a great deal for the new shares.
Now at $74 per share — 20% above IV — each reinvested dividend buys shares at a 20% premium. The DRIP is now taking income you earned and converting it into overpriced shares. Automatically. Every quarter.
The solution is straightforward: when a position reaches or exceeds intrinsic value, turn off DRIP on that position. Take the dividends as cash. Redirect them into something currently undervalued — the next candidate in the pipeline, or a money market fund while you search. Don’t let an automated program make an irrational buying decision on your behalf while you’re not looking.
Where the Money Goes Next
The final step is brief but important.
When you sell a position — because price exceeded IV, because ROIC < WACC, because the thesis broke — you’re not just exiting. You’re freeing capital to redeploy.
The same screening process that found this position finds the next one. The margin of safety discipline that told you to buy Maple Ridge at $42 is scanning for the next Maple Ridge — something currently priced at a 25–35% discount to its intrinsic value, with a clean balance sheet, with cash flows you can model, with a competitive position you can understand and defend.
This is the complete loop: find a business worth more than it’s selling for, buy it at a discount, hold while it compounds, sell when the discipline says the opportunity has closed, and redeploy into the next opportunity the discipline identifies.
The sell is not the end of the process. It’s the handoff. The capital returned from Maple Ridge at $74 becomes the capital invested in the next undervalued business — and the compounding continues, now anchored in a position that actually offers a margin of safety again.
A Note to Luca and Lili
Part 9. The last one.
We’ve come a long way from Part 1, where we asked a simple question: what is a business actually worth? Eight parts later, you know how to answer it — not as a guess, but as a reasoned calculation backed by real numbers.
I started this series for you. I kept writing it for every reader who found it. But I never forgot who I was really talking to.
Here’s what I want you to take from this closing post in particular: the sell discipline is where most investors fail. They learn to be patient buying. They never learn to be disciplined selling. They hold too long because they like the company, or because they don’t know what they’d buy instead, or because selling feels like giving up. Then they watch a 30% gain become a 10% gain become nothing.
The discipline is symmetrical. The same logic that says “I won’t pay more than 80 cents for a dollar” has to also say “I won’t hold past $1.20.” Otherwise the 80 cents was a rationalization, not a principle.
You’ll both outlive any market cycle I’ve seen. Use the framework. Trust the math. And when the position has given you what it promised, let it go — and look for the next one.
— Papa
The One-Sentence Summary
Sell when price exceeds intrinsic value by 20% or more (inverted margin of safety), when the business destroys more value than it creates with each reinvested dollar (ROIC below WACC), or when the assumptions that justified your original estimate have been proven wrong by the business’s actual results — and if DRIP is running, shut it off the moment a position crosses fair value.
This is the final post in the Intrinsic Value series. From here, I’d suggest reading the Growth Investing series — which uses a completely different framework to find value — or continuing with Why Prices Move, which explains what actually drives the gap between price and value in the first place.
— Jim