At the 1998 Berkshire meeting, a shareholder asked how much weight an investor should give to quality versus price. Munger answered in three sentences and effectively closed the subject:
“The investment game always involves considering both quality and price. And the trick is to get more quality than you’re paying for in the price. It’s just that simple. But not easy”
— Charlie Munger
Buffett, immediately: “But not easy.” Munger: “No, but not easy.”
That exchange is funny, and it is also the whole problem. The rule is trivially stateable and almost useless until you can measure both sides of it. Price you can read off a screen. Quality is the hard half — so what is it?
Quality is narrow outcomes, not big growth
Most people, asked to define a high-quality business, reach for growth. Buffett, answering the same question minutes earlier, reached somewhere else entirely:
“…if you talk about quality meaning the certainty that the business will perform as you expect it to perform over a period of time, so the range of possible performance is fairly narrow — you know, that’s the kind of business we like to buy.”
— Warren Buffett
Quality is narrow dispersion of outcomes. Not the height of the expected result — the tightness of the distribution around it.
This matters more than it first appears, because it reframes what you are buying. A business whose earnings in 2035 could plausibly be anywhere between zero and enormous is low quality no matter how thrilling the upside, because you cannot value it and you certainly cannot size it. A dull business earning predictably for the next decade is high quality even if it never excites anyone.
It also quietly disposes of the growth-versus-value argument. Growth is not the opposite of quality and it is not a synonym for it either. Growth is an input to value — sometimes a large one, occasionally a negative one, if it is bought with capital that earns less than it costs.
Takeaway
Ask "how wide is the range of things that could happen to this business?" before you ask "how big could it get?". A narrow range is what makes a valuation meaningful in the first place.Growth alone tells you almost nothing
Buffett has been blunt on this for decades: growth is only good when the capital funding it earns a decent return. A company can grow revenue 15% a year forever and destroy value the entire time, if every dollar of growth requires two dollars of new plant and working capital.
He gave the mechanism its own memorable label in that same 1998 session, warning about businesses that must “spend money like crazy” just to stand still. The tell is simple: profits that never turn into cash, because the cash is immediately consumed by the next round of capital expenditure.
Which is why the useful question is not how fast is it growing but what does growth cost, and who pays for it.
Putting a number on it: Quality Minus Junk
Fat Pitches scores this as QMJ — Quality Minus Junk, 0 to 100. The name and the underlying idea come from the academic factor literature (Asness, Frazzini and Pedersen), and the implementation deliberately stays boring: five metrics, each normalised to a 0–1 range, averaged.
-
Return on invested capital
Does a dollar put into this business come back with friends? The single best test of whether growth is worth having.
ROIC -
Gross profitability
Gross profit relative to total assets. Harder to manipulate than net margin and, historically, a stubbornly good predictor.
GP / A -
Leverage
Net debt against EBITDA. Debt does not make a business bad; it makes a bad year fatal instead of survivable.
ND / EBITDA -
Accruals
The gap between reported earnings and actual cash. A persistent gap is the most polite early warning you will get.
accruals -
Free-cash-flow stability
Buffett's "narrow range", made into a measurement: how steady is the cash across the cycle?
FCF stability
Read the bands as: below 50 weak, 50–75 solid, above 75 exceptional.
Two design choices in there are worth pulling out, because they are where this kind of score usually goes wrong.
It is scored against sector peers, not in the abstract. A software company and a utility have structurally different gross margins and leverage, and comparing them on raw numbers just ranks industries. Where a recognisable peer set exists the anchors are sector-relative; where it does not, absolute anchors are used and that is a known weaker read.
It says nothing about price. By construction. A QMJ of 88 is not a buy signal, it is an answer to a different question — whether this is a business worth wanting. Whether it is worth owning today is settled three steps later, against intrinsic value. Keeping those two apart is the entire discipline; collapse them and you have built a momentum screen with extra steps.
The trap: cheap for a reason
Here is where the Munger line earns its keep, and where screening on price alone fails.
A low multiple is not evidence of a bargain. A business may thoroughly deserve a low multiple, because its economics are deteriorating. A falling price-to-earnings ratio on falling returns is not cheapness — it is the market marking down a melting ice cube slightly faster than you noticed.
The distinction is testable:
Genuinely cheap
The multiple compresses while returns on capital and margins hold. The market's opinion moved; the business did not. This is the setup worth waiting for.
Cheap for a reason
The multiple and the business fall together. Returns, margins and cash conversion are all deteriorating — and it will keep looking cheap all the way down.
This is exactly why quality is judged before price in the app, and never from the same screen. If you meet a company as “a stock trading at 7× earnings”, you have already anchored. If you meet it as “a business with a 6% ROIC and worsening accruals”, the 7× stops looking like an opportunity and starts looking like an explanation.
What it does not do
QMJ is a filter for attention, not a verdict, and the app treats it that way. It will not tell you whether the moat is durable — that is a judgement, and it sits in the conviction checklist where a human has to answer it. It cannot see a regulator, a key-customer concentration, or a founder about to leave. It is backward-looking by construction, because every input is a reported number.
What it does well is cheap triage: it tells you which businesses are worth your time before you spend an evening on a spreadsheet.
Four questions the score cannot answer for you
- Why do customers stay, and what would make them leave?
- Would this business still be earning in ten years if its best competitor doubled its budget?
- Is the recent margin the normal one, or the peak of a cycle?
- Where do I disagree with what the market already believes about it?
Munger’s line survives all of this intact. Quality and price, both, always — and the trick is the gap between them. Measuring the first half just means you have something to compare the second half against.
The full sequence this fits into is in inside the Fat Pitches platform, and the reason quality gets judged before price is in the two risks. The quality score is also what caps the size of a position in our free position size calculator for stocks.
