THE P/E RATIO THAT MADE A BRILLIANT COMPANY LOOK RIDICULOUSLY EXPENSIVE
For years, one number made Amazon look almost impossible to justify as an investment.
That number was the P/E ratio.
The price-to-earnings ratio compares a company's share price with its earnings per share.
A company earning $10 per share with a $100 share price has a:
10× P/E
But Amazon was doing something very unusual.
Its stock price was rising while its reported profits remained relatively small.
That made its P/E ratio look enormous.
In 2015, Amazon's year-end P/E was around 528×. In other words, investors were paying roughly $528 for every $1 of the company's trailing earnings.
To a traditional value investor, that number looked absurd.
But there was a reason.
AMAZON WASN'T TRYING TO MAXIMIZE PROFIT
Amazon's strategy was different.
Jeff Bezos repeatedly told shareholders that Amazon would prioritize long-term market leadership over short-term profitability.
The company's 1997 shareholder letter explicitly said investment decisions would be made with long-term market leadership in mind rather than short-term profitability or Wall Street reactions.
That meant Amazon could make a strange-looking decision:
Earn less today
↓
Spend more on infrastructure, technology and expansion
↓
Build a much larger business
↓
Potentially earn much more later
The accounting statement could therefore make the company look less profitable than its underlying growth story suggested.
THE P/E RATIO WAS LOOKING AT THE WRONG PART OF THE STORY
This doesn't mean P/E was useless.
It was accurately reporting the relationship between Amazon's market price and its earnings.
The problem was that earnings were unusually low relative to the company's growth ambitions.
Amazon was spending heavily on things such as:
The company was deliberately accepting lower short-term profitability in exchange for expansion.
Amazon's 1997 shareholder letter described continued investment in systems and infrastructure as necessary to build customer convenience, selection and service.
THEN LOOK AT THE SALES
This is where Amazon became difficult to analyze using one ratio.
In 1996, Amazon generated about $15.7 million in sales.
By 1997, that had exploded to $147.8 million.
That's approximately 838% growth in one year.
Its cumulative customer accounts increased from about 180,000 to 1.51 million.
Amazon wasn't behaving like a mature company whose main job was to squeeze another percentage point out of its existing business.
It was trying to build something much larger.
THE P/E KEPT LOOKING CRAZY
Amazon's historical P/E illustrates just how strange the numbers became.
Approximate year-end figures:
2010: 70×
2011: 125×
2013: 676×
2015: 528×
2016: 150×
2017: 185×
2018: 73×
2019: 79×
2020: 77×
These numbers come from historical market and earnings data; the ratio became extremely volatile because Amazon's earnings were relatively small compared with its market value.
And sometimes Amazon wasn't profitable enough for a conventional positive P/E to be meaningful at all.
That happened in several periods.
WHY DID INVESTORS KEEP PAYING?
Because they weren't necessarily valuing Amazon based only on what it earned that year.
They were also considering what the company could become.
Amazon had something that traditional P/E analysis struggled to capture:
rapidly expanding businesses inside one company.
Retail was only part of the story.
Then came AWS.
Then advertising.
Then the third-party marketplace.
Then Prime.
Each could change the economics of the overall company.
Amazon itself later described Marketplace, Prime and AWS as three major businesses.
THE AWS EFFECT CHANGED THE ECONOMICS
Amazon Web Services was particularly important.
Traditional retail tends to involve:
Buy inventory
→ store it
→ ship it
→ earn a relatively limited margin.
Cloud computing operates differently.
Customers pay for computing, storage, databases and other services.
Once Amazon had built the infrastructure, additional usage could generate revenue without requiring a traditional retail inventory model for every transaction.
That helped create a business with very different economics from Amazon's original online bookstore.
THEN PROFITS STARTED CATCHING UP
This is the part that makes the P/E story interesting.
A P/E ratio can fall in two ways.
THE STOCK PRICE FALLS
or
EARNINGS GROW
Amazon demonstrated the second mechanism.
As earnings increased, the denominator of the P/E equation became much larger.
So even if the stock price remained high, the P/E could fall dramatically.
Amazon's year-end P/E went from roughly:
528× in 2015
to
150× in 2016
to
185× in 2017
and eventually much lower in later years.
The company didn't need to suddenly become "cheap."
It needed its earnings to grow into its valuation.
THIS IS WHY A HIGH P/E IS NOT AN AUTOMATIC VERDICT
A P/E ratio tells you:
How much investors are paying for each unit of current earnings.
It doesn't tell you by itself:
How quickly earnings are growing
How much the company is reinvesting
Whether margins can expand
Whether new businesses could become important
How durable the competitive advantage is
Whether today's earnings are temporarily depressed
Two companies can both have a 50× P/E and have completely different economic situations.
One might be growing rapidly.
The other might be stagnating.
The ratio alone cannot tell you which one.
BUT THERE WAS REAL RISK
This shouldn't become a story about how everyone who thought Amazon was expensive was simply wrong.
A huge P/E can absolutely signal excessive expectations.
If Amazon had failed to grow its earnings, investors paying hundreds of times earnings could have suffered enormous losses.
The market was effectively demanding extraordinary future performance.
Amazon's own early shareholder letters acknowledged substantial risks, including aggressive competition, execution challenges, geographic expansion and the need for continuing investment.
So the high valuation wasn't free.
It reflected enormous expectations.
THE INTERESTING ACCOUNTING LESSON
Imagine two companies.
COMPANY A
Revenue: $10 billion
Profit: $1 billion
P/E: 20×
COMPANY B
Revenue: $10 billion
Profit: $100 million
P/E: 200×
At first glance, Company B looks dramatically more expensive.
But imagine Company B is deliberately spending hundreds of millions building infrastructure that could eventually produce much larger profits.
The P/E correctly tells you that current earnings are small compared with the market value.
What it doesn't tell you is whether those investments will work.
That's the analyst's job.
AMAZON'S ORIGINAL PHILOSOPHY WAS ALMOST THE OPPOSITE OF "MAXIMIZE THIS YEAR'S P/E"
Bezos' 1997 letter made the philosophy unusually clear.
Amazon said it would judge itself using measures such as:
Customer growth
Revenue growth
Repeat purchases
Brand strength
Market leadership
rather than focusing only on short-term profitability.
That was an unusual philosophy for a public company.
But it explains why Amazon could look financially strange for years.
The company was effectively saying:
"Don't judge the machine only by how much profit it produces today. Watch what we're building."
THE P/E RATIO WASN'T WRONG
This is the important distinction.
The P/E ratio wasn't "fooled."
It was doing exactly what it was designed to do.
Amazon's earnings were small.
Its market value was enormous.
Therefore:
P/E = enormous.
The difficult question was what those earnings would look like several years later.
And that question cannot be answered by the P/E ratio alone.
THE BIGGER LESSON
P/E is most naturally useful when you're comparing companies with reasonably established earnings.
For companies undergoing rapid expansion, heavy reinvestment or major business-model changes, investors often need to examine additional measures:
Amazon's story is a perfect example of why.
A company can look extremely expensive based on today's earnings while investors are actually paying for the possibility of much larger earnings tomorrow.
That possibility can be correct.
Or it can be disastrously wrong.
MAACAT PERSPECTIVE
A P/E ratio answers a very specific question:
"How much am I paying for the company's current earnings?"
It does not answer:
"How much will the company earn five or ten years from now?"
Amazon spent years looking extraordinarily expensive by the first measurement.
Its strategy was to keep investing, expand the business and worry less about maximizing short-term profit.
Eventually, the earnings became large enough to make the earlier P/E numbers look very different in hindsight.
That's the deeper lesson:
A high P/E can mean the stock is overpriced.
But sometimes it means something more complicated:
the market is pricing in a future that hasn't appeared in the earnings statement yet.
