Learn

Why a great business can still lose you money

A company reports record profits and its shares go nowhere for years. That is not a conspiracy or a mistake. It is multiplication, and it takes two minutes to understand for good.

The whole thing in one line

A share price is not a measure of how well a company is doing. It is two separate things multiplied together:

Earnings per share

What the business earned, divided by the shares in issue. This is the company's doing.

The multiple

How many years of those earnings buyers will pay for today. This is not the company's doing at all — it can halve in a year with nothing changing inside the business.

The same thing, actually happening

ResMed passes every structural test in our Exceptional Company Series. Measured from its August 2021 high to August 2026, nothing was wrong with the business — it was one of the best stretches in its history.

Read the last column as multiplication: 3.21 × 0.260 = 0.83. Earnings nearly doubled; buyers decided to pay less than half as much for each dollar of them. The second effect was bigger, so the shareholder lost money — for 60 months.

Two things worth knowing

Why multiples move

Mostly two reasons, neither involving the company: how much growth buyers expect afterwards, and what else they could do with the money. When interest rates rise, a bond starts paying a real return and every share competes with it, so buyers pay less for the same earnings. That appears instantly in the price and never in the results.

It works in reverse, and that one is more dangerous

A mediocre business whose multiple expands from 12× to 24× doubles its share price having earned nothing extra. Its holders will conclude they picked well. They were paid by a change of mood, which cannot be researched or repeated, and can be handed back as fast as it arrived.

How to sense-check a multiple

None of these settle the question alone. Each is a rough cross-check, and read together they tell you whether a price is ordinary or genuinely unusual.

PEG ratio

The P/E divided by the expected earnings growth rate. A P/E of 30 on 30% growth (PEG 1.0) is a different bet from a P/E of 30 on 10% growth (PEG 3.0) — the first pays roughly a year of multiple per point of growth, the second pays three. It only holds up if the growth rate behind it is real, which is the harder question underneath it.

The company's own history

Is today's multiple rich or cheap against where this same business has traded before? Re-rating away from a fifteen-year average is a different event from merely returning to it.

Peers in the same industry

Priced in line with similar businesses, or an outlier? An outlier is not automatically wrong — it is a question that needs an answer, not a red flag on its own.

The market as a whole

Compare against the S&P 500's own average P/E. A premium to the index is a bet that this business deserves to be treated as better than average — sometimes true, always worth stating rather than absorbing by habit.

Analyst targets and consensus growth

Worth a look, not a verdict. Estimates tend to lag price rather than lead it — they get revised up after a stock has already re-rated, not before — so treat consensus growth as one more data point, never the deciding one.

All five are cross-checks on the same underlying question, not a substitute for asking it directly: is the earnings growth priced into this multiple actually going to happen, and for how long? Multiples do not float free — left alone, they drift back toward the industry average, because paying more than that needs a standing reason, not a habit. The reason has to be structural: a moat, a durable trend, evidence the growth is unusually likely to continue. Absent that, today's premium is the thing most likely to revert — which is the same mechanism the de-rating above shows in numbers.

Then why write about businesses at all?

If this is not a recommendation, and a company here might still lose you money for years, what is the work for?

1

The two questions only work in this order. You cannot judge a price until you know what is being priced.

2

Only the first one keeps. Understanding a business takes weeks and stays true for years; a multiple changes in a month.

3

It is what lets you act when the price finally moves. The de-rating above was a company becoming available at half the price to anyone who already understood it — and a frightening chart attached to a stranger, for anyone who did not.

So: exceptional can also be expensive. A wonderful business is a necessary condition for a good long-term investment, and not remotely a sufficient one. Settling the durable half in advance is the point of the series. It is a watchlist built on evidence, not a list of things to buy today.

The follow-up question is how long that takes. A multiple falling on earnings growth alone has a half-life, and it depends only on the growth rate — not on where the multiple started. There is a page for it, with the sliders to work through your own numbers: how long a high multiple takes to come down.

Terms used here are defined in Key terms. Figures: ResMed (RMD), month-end closing prices and trailing twelve-month earnings as announced, from our own store, as at August 2026. General explanation, not personal advice, and not a recommendation to buy or sell anything. Capital is at risk.