The P/E Ratio, Explained: What It Tells You, Where It Fails, and Why SpaceX and Anthropic Break It

Hey everyone,

The P/E ratio is probably the most quoted number in investing. It shows up on every stock page, in every news segment about "expensive" markets, and in nearly every beginner's checklist. It's also under real strain right now, because some of the biggest IPOs in history, SpaceX this summer and Anthropic reportedly next, come with no meaningful P/E at all.

This post covers what the ratio is, what it's good for, where it breaks down, and what investors are reaching for instead when a company has a huge valuation and no profit to divide it by.

What is the P/E ratio?

P/E stands for price-to-earnings. It's the share price divided by the company's earnings per share (EPS) over the same period:

P/E = share price ÷ earnings per share

You get the same answer by dividing the company's total market value by its total net income. Say a stock trades at $100 and the company earned $5 per share over the last year. The P/E is 20. Another way to say it: you're paying $20 for every $1 of yearly profit.

There are a few common versions:

• Trailing P/E uses the last 12 months of reported earnings. It's based on real results, but it looks backward.
• Forward P/E uses analysts' estimates for the next 12 months. It looks ahead, but it's only as good as the forecasts.
• CAPE (Shiller P/E) averages 10 years of inflation-adjusted earnings to smooth out booms and recessions.

What the P/E ratio is good for

Used well, P/E is a fast, rough way to ask "how much am I paying for the profits this business makes?" A few things it does nicely:

• It standardizes price. A $500 stock isn't automatically more expensive than a $20 stock. P/E puts them on the same footing by tying price to profit.
• It lets you compare similar companies. Two grocery chains or two big banks with different P/Es are worth a closer look. Why is one cheaper? Is it slower-growing, riskier, or just overlooked?
• It shows what the market expects. A high P/E generally means investors expect earnings to grow quickly. A low one often signals slow growth, higher risk, or pessimism.
• It gives you a sense of the whole market. Comparing today's market P/E to history is a quick check on whether stocks, as a group, are priced richly or cheaply.
• It converts to an earnings yield. Flip it: 1 ÷ P/E. A P/E of 20 is a 5% earnings yield, which you can hold up against what a Treasury bond pays.

S&P 500 trailing P/E

A good rule of thumb: P/E is a starting question, not an answer. It tells you where to dig, not whether to buy.

Where the P/E ratio falls short

P/E has real blind spots, and most of them matter more now than they used to.

• It doesn't exist without profits. If earnings are negative, the ratio is meaningless. Money-losing companies just get an "N/A", which tells you nothing about whether the price is reasonable.
• Earnings can be noisy or distorted. One-time gains, write-downs, and accounting charges swing net income a lot. Anthropic's reported 2025 net loss of roughly $42 billion, for example, included about $34 billion of non-cash accounting charges (more on that in our breakdown of its leaked prospectus). A P/E built on that number would say almost nothing about the actual business.
• It ignores growth. A P/E of 40 can be a bargain for a company doubling profits every year and a rip-off for one growing 2%. That's why investors pair it with growth, often through the PEG ratio (P/E divided by expected earnings growth).
• It ignores debt. Two companies can have the same P/E but very different balance sheets. Enterprise-value measures like EV/EBITDA account for debt and cash. P/E doesn't.
• It's cyclical. Earnings at a cyclical company (airlines, chip makers, homebuilders) peak in good times, which makes the P/E look low right when risk is highest, and look high at the bottom when it may be a better entry.
• It depends on the industry. A software company, a utility, and a bank have structurally different P/Es. Comparing across sectors misleads.
• Companies can manage it. Buybacks shrink the share count and lift EPS without any improvement in the business, and accounting choices can nudge reported earnings.
• It's backward-looking by default. Trailing earnings describe the past. Stocks are priced on the future.

Enter the pre-profit mega-IPO

Traditionally, a company went public after it had a record of profits to price off. That's changing. Companies are staying private longer, raising enormous sums from private investors, and arriving on the public market at valuations in the hundreds of billions or trillions, often before they've earned a dollar in net income.

SpaceX is the clearest example so far. It priced its IPO on June 12, 2026 at $135 per share, a valuation of roughly $1.77 trillion. Its S-1 showed about $18.7 billion of 2025 revenue and a net loss of $4.9 billion (an operating loss of about $2.6 billion, though adjusted EBITDA was about $6.6 billion). Starlink was around 61% of revenue. With negative earnings, SpaceX has no P/E to speak of.

Anthropic would be the next. As we covered in our post on its leaked draft prospectus, backers reportedly expect a valuation above $2 trillion against a $4.6 billion 2025 revenue base and a $65 billion annualized run rate by mid-2026. Again, a huge price with no usable P/E.

Price-to-sales at the headline valuation

What investors use instead

When there are no earnings to divide by, investors move up the income statement or stretch the timeline:

• Price-to-sales or EV/revenue. It works without profits, but it ignores margins entirely. A dollar of revenue at a 70% gross margin is worth far more than one at 10%.
• Run-rate revenue. Annualizing the latest quarter or month captures fast growth, but it assumes the surge holds.
• Unit economics and margins. Gross margin, customer retention, and revenue per customer (SpaceX's reported Starlink revenue per user, for instance, fell from about $99 to $66 a month as it expanded) show whether growth is healthy.
• Adjusted EBITDA and free cash flow. Adjusted EBITDA is handy, but it leaves out capital spending. SpaceX reported about $6.6 billion of adjusted EBITDA alongside a net loss, and heavy infrastructure spending.
• Forward or "future" P/E. Estimate earnings several years out, assume a normal margin, and ask what multiple you'd be paying then.

That last one is worth a quick exercise. At a market-like P/E of 20, a $2 trillion valuation needs about $100 billion of annual net income to be "fairly" priced. If a company earned a 25% net margin, that implies about $400 billion of revenue. For $1.77 trillion, the same math points to roughly $88 billion in earnings, or around $350 billion of revenue at a 25% margin, compared with $18.7 billion in 2025. This is our back-of-the-envelope math, not a forecast, but it shows how much growth is baked into the price.

Is the P/E ratio becoming obsolete?

No, but its job is changing. For mature, profitable businesses, which still make up most of the market, P/E remains a useful shorthand. For pre-profit giants, it's replaced by a longer chain of questions: how fast is revenue growing, will margins expand, how much capital does growth require, and when do profits actually show up?

Two risks stand out. First, sales multiples can make anything look reasonable if you assume growth lasts long enough. Second, when a big share of future profit is already priced in, small disappointments can move the stock a lot. SpaceX is a live example: after a spike to an all-time high of $225.64 within days of its debut, the stock fell sharply after its lock-up expiry and first earnings report, and has recently traded near $150, still above the $135 IPO price.

The takeaway isn't "ignore P/E" or "P/E is dead." It's that no single multiple tells the whole story, and the less profit a company has, the more you're paying for a forecast rather than a track record.

What to take from it

• P/E tells you the price of a dollar of profit. It's quick and useful for comparing similar, profitable companies.
• It's a starting point. Pair it with growth, debt, margins, and cash flow before drawing conclusions.
• No earnings, no P/E. For pre-profit companies, you're judging a story, so check the assumptions behind it.
• Know what's in the price. The higher the multiple, the more perfect the future has to be.

If you want to go deeper on valuation, our free DCF course shows how to value a company on its future cash flows, and the ratios course walks through P/E and its companions in more detail.

More soon,
Learn to Love Money

Sources

• Multpl: S&P 500 P/E ratio
• Via Satellite: SpaceX's IPO filing gives first look into company's financials
• SEC: Space Exploration Technologies S-1
• Wikipedia: Initial public offering of SpaceX
• Fortune: Anthropic's $2 trillion IPO prospectus has leaked
• PYMNTS: Anthropic's IPO filing puts a $518 billion price tag on AI ambition