What a company is worth and what affects it

Multiples are shortcuts for a hard question: whether a company is expensive or cheap. They work reasonably well comparing similar companies, and not at all comparing different businesses. The common error isn't calculating them wrong: it's comparing them wrong.

What the P/E ratio measures and why it's misread

The P/E ratio — price to earnings — divides a share's price by the profit it generates per share. A P/E of 15 means you're paying fifteen euros for every euro of annual profit.

The correct reading, and the incorrect one

Put the other way it's clearer: if the company kept earning exactly the same every year and handed you the whole profit, it would take fifteen years to recover what you put in.

From that comes the intuitive interpretation — low P/E, cheap; high P/E, expensive — and from that comes the most common error in retail analysis.

A low P/E can mean two opposite things: that the market is undervaluing a good company, or that the market expects its profits to fall. The second is considerably more frequent.

And a high P/E can too: that the stock is expensive, or that strong growth is expected which justifies the price.

The isolated number doesn't distinguish between those. It only says what you're paying for current profit, and price forms on future profit.

The low-P/E trap

It's the most expensive one and has its own name in the industry: the value trap.

A company whose business is deteriorating trades at a P/E of 6 because the market already knows next year's profit will be half. Against that future profit, the real P/E isn't 6: it's 12. The stock was never cheap, it was anticipating a fall.

It's exactly the same mechanism as the high-dividend trap we saw in the corresponding guide: a very attractive ratio is usually attractive because the denominator is about to worsen, not because the numerator is a bargain.

Trailing and forward P/E aren't the same number

Here's a distinction almost no website flags and that completely changes the interpretation. When you see "P/E 18" somewhere, it can be calculated two different ways.

Trailing P/E, which uses the last twelve months of published profit. It's a verified fact: that profit existed.

Forward P/E, which uses estimated profit for the next twelve months. It's a forecast, built by the same analyst consensus we covered in the earnings chapter — with all its limitations.

Both are called the P/E and they use different profits: one already published, one estimated. That difference makes the same number come out higher or lower depending on where the company is in its life, which is why comparing one against the other compares nothing.

	Trailing P/E	Forward P/E
Which profit it uses	The last twelve months, already published	The estimate for the next twelve months
Where that figure comes from	The accounts filed with the regulator	Analyst consensus
How certain it is	A verified fact: that profit existed	A forecast, with all its uncertainty
Where it comes out lower	In a company whose profit is falling	In a company whose profit is growing
What it can be compared against	Another company's trailing P/E	Another company's forward P/E

Why the difference matters

Compare one company's trailing P/E against another's forward P/E and you're comparing nothing. And it happens constantly, because different sources publish different calculations without specifying which they use.

The basic check before using any multiple: know whether the profit in the denominator is real or estimated. A multiple built on an estimate inherits all that estimate's uncertainty.

What EV/EBITDA adds

The P/E has a large blind spot: it ignores debt entirely. Two identical companies, one debt-free and one loaded with it, can show the same P/E. EV/EBITDA (enterprise value divided by EBITDA: earnings before interest, tax, depreciation, and amortisation) exists to fix that.

What enterprise value is, with a house

Enterprise value is the best way to understand what you're really buying, and it's understood through an example with nothing financial about it.

You see a house advertised at €200,000. Before signing you discover two things: it carries a €150,000 mortgage that becomes yours to pay, and inside there's a safe holding €50,000 in cash that becomes yours.

What does that house genuinely cost you, clean of everything?

€200,000 you pay, plus €150,000 of mortgage you assume, minus €50,000 you find inside: €300,000.

With a company it's identical. Market capitalisation — price per share times number of shares — is the advertisement. Enterprise value adds the debt and subtracts the cash, answering the real question: what it would cost to own the whole business.

Looking only at market cap is buying a house without asking whether it has a mortgage.

And EBITDA in the denominator

EBITDA is profit before interest, tax, and depreciation. It's used because, by not deducting interest, it allows companies with different debt loads to be compared on the same basis: the numerator already includes the debt, so the denominator shouldn't penalise it again.

Its limitation is worth stating: by excluding depreciation, EBITDA ignores that factories and machinery wear out and need replacing. In an asset-heavy business, EBITDA presents a rosier picture than reality. Which is why the free cash flow from the previous session remains the most honest number of all.

Why a multiple only works within the same sector

Because each sector has a normal range, and that range reflects real business differences: how fast it grows, how much investment it needs, and how predictable its revenue is.

Sector type	Typical EV/EBITDA	Why
Utilities	6-10 times	Highly predictable revenue, low growth, heavy network investment
Consumer staples	10-14 times	Stable demand, moderate growth
Technology and biotech	Frequently above 15	High expected growth, little need for physical assets

Benchmarks compiled by Valutico, a firm specialising in valuation methodology.

What this means and what it doesn't

Technology trading at 18 times and a utility at 8 doesn't mean technology is expensive and the utility is cheap. It means they're different businesses and the market pays differently for each type of earnings stream.

Comparing a utility against another utility does inform. Comparing a utility against a tech company informs nothing, in the same way comparing the price per square metre of a city-centre flat with an industrial warehouse doesn't say which is the better buy.

The two comparisons that do work

They're the same two from this chapter's first session, and that isn't coincidence: against direct competitors and against its own history.

The second is the most forgotten and among the most useful. A company that has always traded between 12 and 16 times and now sits at 9 raises a specific question: has the business changed, or has the market's mood changed? Those are very different answers, and the accounts help decide which.

Why price history changes after a split

A stock split multiplies the number of shares and divides the price in the same proportion. Your position is worth exactly the same, but if the price history isn't adjusted, the chart shows a fall that never happened.

On October 30, 2025, Netflix filed with the SEC (Securities and Exchange Commission), the US securities regulator, for a ten-for-one split. Every shareholder of record as of November 10 received nine additional shares for each one held, and from November 17 the stock traded at roughly a tenth of the previous price.

Someone holding one share worth around $1,000 ended up with ten worth around $100. Their position was worth exactly the same the day before and the day after.

Why this breaks a chart

Picture that chart unadjusted. On split day, the price would go from $1,000 to $100: a 90% fall that cost nobody a cent.

Any support or resistance level calculated on that history would be useless from that date. Any moving average, any volatility calculation, any comparison against previous highs: all broken.

That's why serious platforms rewrite the entire prior history, multiplying it by the same factor, so the chart tells one continuous story.

Three separate operations change how a price history reads without necessarily changing what your position is worth. All three require the chart to be rewritten backwards, and all three get mistaken for falls or rises that never happened if nobody does it.

	Stock split	Share issue	Spin-off
What the company does	Divides each share into several	Issues new shares to raise money	Separates part of the business into an independent company
What happens to you	You hold more shares, each worth less	You hold the same shares, out of a larger total	Your position becomes two
What happens to your share of the company	Unchanged	It shrinks: that's dilution	It splits across two companies
What happens to an unadjusted chart	A fall appears, the size of the split	A jump appears matching no news	The history stops corresponding to the remaining business

All three share the same practical problem: they change how the history reads without necessarily changing the value of what you hold.

And here's a gap that isn't theoretical. Adjusting history for a split looks trivial until it's done wrong, and doing it wrong produces price levels that never existed. It's the kind of silent error that breaks nothing visibly: it simply makes every subsequent calculation wrong.

Volatly rebuilt part of its own engine after detecting exactly that problem, and today treats each corporate action as an adjustment applied one way only, from a single source. It's invisible work and without it nothing built on top of it means anything.

Which sectors suffer when rates rise

Sectors don't react identically to a change in rates, and the difference isn't that some are better: it's whether their demand depends on people feeling flush. That dependence explains a good share of the moves that look inexplicable.

	Cyclical	Defensive
What they sell	Things bought when confidence is high and credit is cheap: cars, travel, renovations, technology	Things bought regardless: basic food, electricity, medicine
How their demand behaves	Rises and falls with the economic cycle	Barely shifts with the cycle
What happens when rates rise	Their customer postpones the purchase and their debt gets more expensive	Demand holds up, but their dividend competes with bonds now paying more
Example sectors	Automotive, leisure, real estate, industrials, technology	Food, utilities, pharmaceuticals

Neither column is the good one. They're two different ways of depending on the economy, and which affects you more depends entirely on what you hold and when you need it.

The two mechanisms by which rates do damage

One: debt gets more expensive. That's what we saw in the previous session. A heavily indebted company refinancing at a higher rate sees its interest bill rise without doing anything differently.

Two, and this is the least-explained one: bonds compete. When rates rise, government debt and bonds start paying attractive interest with barely any risk. An investor who accepted a stock with a 4% dividend yield because there was no alternative can now get 4% from a government bond.

That second mechanism hits hardest at sectors bought precisely for their steady dividend: utilities, listed property, telecoms. Their businesses aren't getting worse — their relative appeal is shrinking.

Why real estate takes the double hit

It combines both mechanisms at once. Companies in the sector are usually heavily indebted to finance properties, so they pay more on their debt. And their main appeal to investors is the dividend, which now competes with bonds yielding more.

On top of that sits a third effect on the business itself: if mortgages get more expensive, fewer properties sell and prices fall.

What this means for reading a move

When you see a company fall without having published anything, the cause may not be in the company. It may be in an inflation figure, a central bank decision, or a move in bonds.

Telling a move with a business cause from one with a macroeconomic cause is among the things that saves the most grief, because they're two situations demanding opposite responses: one questions the case for the company and the other doesn't.

What to remember

• A low P/E can mean the stock is cheap or that profit is about to fall; the number alone doesn't distinguish.
• Enterprise value adds debt and subtracts cash: looking only at market cap is buying a house without asking about the mortgage.
• Every sector has its normal multiple range, so they only inform compared within the same sector and against the company's own history.
• Rates hit through two channels: they make debt more expensive and they make bonds compete with dividend stocks.

Milestone reached

That closes Chapter 5. You can open a quarterly report and know where to look: the five lines of the income statement, what each margin measures, what the balance sheet shows that the income statement doesn't, why profit and cash don't match, and what multiples do and don't say.

Chapter 6 changes tools entirely: it leaves the accounts and moves to the chart, starting with the uncomfortable question of whether technical analysis actually works.

Related: the dividend, start to finish · what "risk" really means · how to read a full event