A stock trades at $150 — is that cheap or expensive? The answer determines whether you make money long-term or lose it. Price alone tells you nothing: a Berkshire Hathaway share at $600,000 can be a bargain, a Nvidia share at $100 can be overpriced. Valuation methods — P/E, P/CF, and P/B — translate the share price into a comparable number that shows what you’re actually paying per unit of earnings, cash flow, or book value. This guide explains the three most important multiples, when each one makes sense, and where the typical traps lie.

Why do you need valuation methods?

The absolute share price is an arbitrary number. How many shares a company issues is a design decision — Apple and Tesla have run stock splits, Berkshire Hathaway never has. A stock at $50 therefore says nothing about valuation. What matters is the relationship between the market price and the company’s economic output.

That’s exactly what valuation multiples deliver: they put price into context with earnings, cash flow, revenue, or book value, making companies comparable. If Apple has a P/E of 30 and Microsoft a P/E of 32, you can say “Apple is valued slightly cheaper relative to Microsoft.” Without these multiples, you’d have to value stocks by gut feeling or charts — not a solid foundation for long-term investing.

A second benefit: multiples show how the market values a stock relative to its own past. If the current P/E sits well above the 10-year average, that signals high expectations — either the stock is overpriced or the business model has structurally improved. No other metric delivers that context as compactly. But valuation multiples aren’t magic truth machines — they’re tools that only work in the right combination.

Price-to-Earnings Ratio (P/E)

The price-to-earnings ratio is the best-known valuation metric in finance. The formula: P/E = share price ÷ earnings per share (EPS). The aggregate version: P/E = market capitalization ÷ annual net income. Both produce the same result.

The interpretation: the P/E tells you how many years of earnings you’re paying at the current price. A P/E of 20 means — assuming flat earnings — it would take 20 years for your investment to pay back through profits. Low P/Es are considered “cheap,” high ones “expensive.” But that simple rule has limits: growth companies earn their valuation through future earnings increases, not today’s profits.

Industry averages (as of 2026, US market):

• Utilities and banks: 8 to 14 — limited growth, high stability, often strong dividend yields.
• Consumer staples and industrials: 15 to 22 — moderate growth, established business models.
• Established tech: 25 to 40 — Microsoft, Apple, Alphabet sit here.
• Hyper-growth (AI, cloud): 40 to 80+ — Nvidia, Tesla, smaller cloud specialists.
• S&P 500 overall: historically 15 to 18, currently around 22 to 24.

You compare P/E ratios only within the same industry. Comparing a bank with a P/E of 9 to Nvidia with a P/E of 50 is meaningless — the business models and growth rates are completely different. A useful comparison is bank-to-bank: if Deutsche Bank trades at a P/E of 6 and JPMorgan at 12, look into the reasons — usually it comes down to risk profile, return on equity, or market position.

Price-to-Cash-Flow Ratio (P/CF)

The P/CF ratio puts share price in relation to operating cash flow per share rather than earnings: P/CF = share price ÷ operating cash flow per share. The crucial difference: cash flow shows how much actual cash a company generates, while earnings can be distorted by accounting effects.

When is P/CF better than P/E? In three situations:

First, in cyclical industries. Automakers, steel, and chemical companies have extreme earnings swings. In boom years the P/E looks low and cheap; in crisis years it spikes or turns negative. Cash flow is usually more stable — it reacts less to one-off write-downs and restructuring charges. If you’re valuing BMW or ArcelorMittal, keep an eye on cash flow.

Second, for capital-intensive business models. Telecom carriers, oil and gas companies, and real-estate REITs carry heavy depreciation on their infrastructure. That depreciation suppresses accounting earnings but isn’t a real cash outflow — the money was spent years ago to build a pipeline or mobile network. A REIT at a P/E of 35 looks expensive but often has a P/CF of 12 — and that’s the more honest figure.

Third, for companies that smooth their earnings. When a company reports suspiciously consistent quarterly profit growth, compare it to cash flow. A wide gap (earnings rising, cash flow flat) is a warning sign — revenue may be booked too aggressively or costs deferred. Classics like Enron and Wirecard would have been easier to spot through cash flow than through P/E.

Industry averages for P/CF tend to run 30 to 40 percent below the P/E for the same sector, because cash flow is higher than reported earnings (depreciation isn’t deducted). A P/CF below 10 is generally considered cheap across industries; above 25 is expensive.

Price-to-Book Ratio (P/B)

The P/B ratio compares share price to book value of equity per share: P/B = share price ÷ book value per share. Book value is the sum of all assets minus liabilities — what the company would theoretically be worth on paper if liquidated today.

A P/B of 1.0 means you’re paying exactly book value. Below that (say, 0.7), you’re buying the company at a discount — the market doesn’t trust the book value, or the stock is undervalued. Above (say, 8.0), you’re paying eight times book value, which is only justified by brand strength, patents, or future earnings.

P/B is especially relevant for banks and insurers. These business models consist at their core of financial assets — loans, bonds, securities — that sit on the balance sheet at market values or amortized cost. The book value here is a very realistic estimate of company worth. At a bank with a P/B of 0.5, you’re theoretically buying a dollar of equity for 50 cents — that can be an opportunity, or a hint at hidden risks (non-performing loans, lawsuits, regulatory trouble).

For tech companies, P/B is almost useless. Microsoft has a book value per share of around $35 against a share price of $460 — a P/B of 13. That’s not because Microsoft is overpriced, but because the real value (cloud market share, lock-in effects, brand power) doesn’t sit on the balance sheet. Investments in software, R&D, and people are expensed immediately, not capitalized. Comparing P/B between Microsoft and a steelmaker is therefore nonsense.

Rule of thumb: P/B is useful for banks (typical 0.7 to 1.5), insurers (0.8 to 1.8), and asset-heavy industrials (1.0 to 3.0). It’s not useful for software, platform, or brand-driven companies.

Forward vs. Trailing Multiples

Every multiple has two time variants: trailing (backward-looking) and forward (forward-looking). The difference often determines how you read a stock.

Trailing multiples use the last 12 months. Example: trailing P/E = current price ÷ EPS over the past four quarters. Advantage: these numbers are real and documented in financial statements. Disadvantage: they describe the past. A company that just exited a tough year often looks “expensive” on a trailing P/E even though the next years will look much better.

Forward multiples use analyst estimates for the next 12 months. Example: forward P/E = current price ÷ expected EPS over the next four quarters. Advantage: it values the future, which is what you’re actually buying. Disadvantage: estimates can be wrong — and analysts have a historical bias toward optimism, especially with growth stocks.

In practice, the forward P/E sits noticeably below the trailing P/E because analysts expect earnings to rise. For Nvidia in 2024–2025, the trailing P/E often sat at 80 to 100 while the forward P/E was 35 to 45 — the market was pricing in massive earnings growth. Always compare like with like: trailing to trailing, forward to forward. Mixing leads to wrong conclusions.

Pro tip: calculate the PEG ratio (P/E ÷ expected earnings growth rate in percent). A PEG below 1 suggests a fairly valued growth stock, above 2 indicates high expectations. This metric normalizes the P/E by growth.

Worked example: Apple, Nvidia, Berkshire Hathaway

Three companies, three completely different valuation profiles. As of April 2026, figures rounded for clarity:

Apple (consumer tech, established). Market cap roughly $3.4 trillion, net income $100 billion, operating cash flow $115 billion, equity $70 billion.

• P/E: 34 — above Apple’s historical average (around 18), but normal in tech.
• P/CF: 30 — strong cash generation, slightly higher than earnings (typical for tech).
• P/B: 49 — very high, because massive share buybacks have shrunk the equity base.

Nvidia (AI hyper-growth). Market cap roughly $3.2 trillion, net income $75 billion, operating cash flow $80 billion, equity $65 billion.

• P/E: 43 — high, but lower than 2024 (when it ran above 70). The valuation grew alongside earnings.
• P/CF: 40 — similar to P/E, strong cash generation.
• P/B: 49 — very high like Apple, but for a different reason: Nvidia’s real assets are GPU architecture and the CUDA ecosystem, not physical plant.

Berkshire Hathaway (conglomerate, value). Market cap roughly $1.1 trillion, net income $95 billion (highly variable due to mark-to-market on its equity portfolio), operating cash flow $50 billion, equity $600 billion.

• P/E: 12 — visually very cheap, but distorted by mark-to-market swings on the stock portfolio.
• P/CF: 22 — more realistic than the P/E, but hard to compare to peers given the unusual model.
• P/B: 1.8 — low, because Berkshire carries real assets (insurance reserves, fixed assets, equity portfolio) on its balance sheet. Buffett himself uses P/B as the central valuation metric for his own company.

The takeaway: three radically different profiles. Apple and Nvidia look similar at first glance but are valued on completely different growth assumptions. Berkshire is practically built for P/B analysis, while P/B carries no useful signal for Apple or Nvidia. Anyone who only compares the three on P/E misses the structural differences.

Which method when? A decision flowchart

Choosing the right valuation method depends on the company type. Follow this decision tree:

Looking at all multiples simultaneously and putting them in industry context lets you see stocks from multiple angles and avoid one-sided conclusions. As a complement, browse stock profiles in our database, which show all three multiples side by side.

Common Mistakes in Valuation

Deepen your knowledge in the glossary

To go deeper, the BMInsider Finance Glossary covers all the key valuation terms: P/E Ratio, Book Value, Market Capitalization, Free Cash Flow, Net Income, Dividend, and Fundamental Analysis. Current multiples for individual companies are always available in our stock database.

Frequently Asked Questions on Stock Valuation

What is a good P/E ratio? There’s no universally good P/E value — it depends on industry and growth stage. Banks and utilities typically run between 8 and 14, classical industrials between 15 and 22, tech companies between 25 and 40. Growth stocks can justify P/Es above 50 if earnings growth is correspondingly high. More important than the absolute number is the comparison to industry peers and to the company’s own historical average.

When is P/CF better than P/E? P/CF is superior in cyclical industries (autos, steel, chemicals), capital-intensive business models (telecom, oil, REITs), and companies with large one-off write-downs or restructuring costs. It’s more honest than P/E because it isn’t distorted by accounting effects — cash flow is what the company actually generates in real money.

Why is P/B so important for banks? Banks economically consist of financial assets (loans, bonds, securities) that sit on the balance sheet at near-real values. Book value is therefore a realistic estimate of company worth. A P/B below 1.0 signals that the market doesn’t trust the book value — either an opportunity or a warning of hidden risk.

What’s the difference between trailing and forward P/E? Trailing P/E uses actual earnings from the past 12 months; forward P/E uses analyst estimates for the next 12 months. Trailing values are real but backward-looking. Forward values are forward-looking but exposed to estimate error. For growth stocks, forward P/E typically sits well below the trailing P/E.

Can the P/E be negative? Yes — when a company loses money, earnings are negative and the P/E is technically negative. In practice it’s reported as “n/a” because it carries no meaningful economic signal. For loss-making growth stocks (early-stage Tesla, Amazon, Uber), analysts use price-to-sales or forward P/Es on expected future earnings instead.

Which valuation method does Warren Buffett use? Buffett relies primarily on P/B — and on his own estimate of “intrinsic value.” In Berkshire’s annual reports he tracks book value per share over 50 years as the central performance metric. For external analysis he combines P/B, normalized P/E (smoothed over multiple years), and an understanding of the business model. Multiples alone aren’t enough for him — he wants to understand the company.

This article is part of our Academy series — investment education for beginners and beyond.

Disclaimer: This article is for educational and informational purposes only and does not constitute investment advice. The multiples shown for Apple, Nvidia and Berkshire Hathaway are point-in-time values rounded for clarity — actual numbers move daily. Consult the official annual reports for exact figures and speak with a qualified financial advisor before making investment decisions.