Why one ratio is never enough

A mechanic does not use a single tool for every repair, and an investor should not use a single ratio for every company. P/E, P/B and EV/EBITDA are three of the most widely quoted valuation ratios, and each one answers a different question. Used on the right kind of business, a ratio can give a quick and useful read on whether a price looks reasonable. Used on the wrong kind, it can be misleading, and sometimes plainly meaningless.

The purpose of this article is to explain what each ratio measures, which businesses it suits, where it breaks down, and how to choose between them. A valuation ratio is a starting point for research, not a verdict. It tells you what the market is currently paying, not whether that price is right.

What a valuation ratio actually does

Every valuation ratio compares a price with something that represents the business: its profit, its assets or its operating earnings. The result is a multiple, which tells you how much investors are paying for each rupee of that underlying measure. A higher multiple means investors are paying more for the same profit or assets, usually because they expect faster growth, higher returns or lower risk. A lower multiple means the opposite, or sometimes that the market sees problems that are not yet visible in the numbers.

The choice of ratio therefore depends on what drives value in a particular business. For a company that earns steady profits, earnings are the natural yardstick. For a bank, the balance sheet is the business. For a company with heavy borrowing and large machinery, you need a measure that sees through its financing and its capital structure.

P/E: the price you pay for each rupee of profit

The price-to-earnings ratio divides the share price by earnings per share. Equivalently, it divides the company’s market value by its net profit. If a share trades at Rs. 500 and the company earns Rs. 25 per share, the P/E is 20. Put simply, investors are paying Rs. 20 for every Re. 1 of annual profit. Its inverse, called the earnings yield, is 5% in this case, which gives a rough sense of the profit return on the price paid before any growth.

P/E comes in two common forms. Trailing P/E uses the past twelve months of reported earnings, while forward P/E uses an estimate of the next twelve months. Trailing figures are factual but backward-looking, and forward figures are forward-looking but depend on estimates that can be wrong. It is useful to know which one you are looking at before comparing numbers.

P/E works best for companies with stable, positive and reasonably predictable profits. This describes many consumer goods companies, large IT services firms, pharmaceutical companies and other established businesses where earnings are a good guide to the value created. In these cases P/E is easy to compare across companies and across time.

P/E also has well-known limits. It is meaningless when a company makes a loss, because a negative or near-zero denominator produces a figure that tells you nothing. It can mislead for cyclical businesses, where earnings are very high at the peak of a cycle, which makes P/E look low just before profits fall, and very low at the bottom, which makes P/E look high just before profits recover. Net profit can also be distorted by one-time items such as asset sales, write-offs or tax adjustments, so it is wise to look at underlying earnings. And P/E ignores debt: two companies can have the same P/E while one carries far more borrowing and therefore more risk.

Finally, P/E says nothing on its own about growth. A P/E of 30 may be reasonable for a company growing profits quickly and unreasonable for one growing slowly. Some investors use the PEG ratio, which divides P/E by the expected growth rate, as a rough way to put growth beside price, though it depends heavily on a growth estimate that is uncertain.

P/B: the price you pay for each rupee of net worth

The price-to-book ratio divides the share price by book value per share, which is the company’s net worth (assets minus liabilities) divided by the number of shares. If a bank has a book value of Rs. 200 per share and the stock trades at Rs. 300, the P/B is 1.5. Investors are paying Rs. 1.50 for every Re. 1 of the accounting value of shareholders’ equity.

P/B is most useful for businesses whose balance sheet is the engine of value: banks, non-banking finance companies, insurers in some contexts, and asset-heavy businesses such as real estate or certain holding companies. A bank’s earnings come from lending its capital and borrowed money, so its net worth is a meaningful anchor for value.

P/B should almost always be read alongside return on equity (ROE). A bank that earns a high return on its equity deserves a higher P/B, because each rupee of book value produces more profit. A bank earning a return below what investors require should trade closer to or below book value. This is why a high P/B is not automatically expensive and a low P/B is not automatically cheap: the ratio only makes sense when matched with the quality of returns and the quality of the assets.

The limits of P/B are worth knowing. Book value is based on historical accounting, so it can differ greatly from the true worth of assets. For lenders, the quality of the loan book is critical: if a meaningful share of loans turn bad and provisions are inadequate, the stated book value may be overstated. For businesses that rely on brands, software, talent or intellectual property, much of the real value is not recorded on the balance sheet, so P/B is usually high and not very informative. Share buybacks and large write-offs can also change book value in ways that distort comparisons over time.

EV/EBITDA: valuing the whole business, regardless of how it is financed

Enterprise value, or EV, represents the price of the entire business, not only the equity. It is calculated as market capitalisation plus debt (and any minority interest and preference capital), minus cash and equivalents. It answers the question of what it would cost to buy the whole operation, including taking on its debt, net of the cash you would receive. EBITDA stands for earnings before interest, taxes, depreciation and amortisation, which is a rough measure of operating earnings before financing and accounting choices.

Suppose a company has a market capitalisation of Rs. 1,000 crores, debt of Rs. 400 crores and cash of Rs. 100 crores. Its enterprise value is Rs. 1,300 crores. If its EBITDA is Rs. 200 crores, then EV/EBITDA is 6.5.

The great advantage of this ratio is that it is neutral to capital structure. Two companies in the same industry can have very different levels of debt, interest costs and tax rates, which makes their P/E ratios hard to compare. EV/EBITDA puts them on a more level footing by looking at operating earnings relative to the whole enterprise. This is why the ratio is widely used for capital-intensive and debt-heavy businesses such as telecom, infrastructure, power, cement, steel, oil and gas and hospitals, and in acquisitions, where a buyer takes on the whole business including its debt.

The ratio has important limits too. EBITDA ignores depreciation and capital expenditure, so it can flatter a business that must keep spending heavily on machinery or networks just to stand still. Two companies with the same EBITDA may have very different free cash flow once the spending needed to sustain them is counted. The ratio is generally unsuitable for banks and financial companies, because interest is part of their core operations, not a financing cost to be stripped out. EBITDA is also not tightly defined, and companies sometimes report adjusted versions, so it is worth checking exactly what has been added back. Lease accounting can make comparisons tricky as well. In practice, investors often look at EV/EBITDA together with measures such as EV/EBIT, which accounts for depreciation, or free cash flow yield.

Matching the ratio to the kind of company

For stable, profitable businesses with predictable earnings, such as established consumer goods, IT services and pharmaceutical companies, P/E is usually the natural starting point, supported by growth and return measures such as ROE and ROCE. For banks, non-banking lenders and many financial companies, P/B read together with ROE and asset quality is the standard approach, with P/E as a secondary check. Insurers are often analysed with specialised measures as well, which are beyond the scope of this article.

For capital-intensive and debt-heavy companies such as telecom, power, infrastructure, cement and large manufacturers, EV/EBITDA is a common choice, ideally cross-checked with capital expenditure, debt levels and cash flow. For cyclical businesses such as metals, commodities and some auto and capital goods companies, no single year’s ratio is reliable. It is usually better to use earnings or EBITDA averaged across a full cycle, or to compare against the company’s own historical range through good and bad years.

For real estate developers and holding companies, price relative to estimated net asset value, or a sum-of-the-parts approach that values each business separately, is often more informative than a simple P/E. For early-stage or loss-making growth businesses, analysts sometimes turn to price-to-sales or EV-to-sales together with an assessment of the path to profit, while recognizing how uncertain those valuations are.

A hypothetical illustration: same P/E, different risk

To see why one ratio can hide important differences, consider two made-up companies, Alpha Ltd and Beta Ltd. Each has a market capitalisation of Rs. 1,000 crores and earns a net profit of Rs. 50 crores, so both trade at a P/E of 20.

Alpha has no debt and holds a modest amount of cash. Beta, however, has net debt of Rs. 800 crores. On P/E alone they look identical. But the enterprise value tells a different story: Alpha’s EV is roughly Rs. 1,000 crores, while Beta’s EV is about Rs. 1,800 crores, which shows that Beta’s shareholders sit on top of a much larger obligation. If Beta’s operating earnings fell, its interest costs would still have to be paid, and its equity would be hit much harder than Alpha’s. Looking at interest coverage, meaning operating profit divided by interest expense, would show how comfortable Beta really is.

The lesson is that P/E measures what shareholders pay for profit, but it does not show how much borrowing stands in front of them. That is why experienced analysts rarely rely on one ratio alone.

Context matters more than the number

A ratio only becomes useful when it is put in context. Compare it with the company’s own history, because a stock trading well above or below its usual range deserves an explanation. Compare it with direct peers in the same sector, because different industries have very different normal ranges and comparing a bank’s P/B with a software company’s P/B is meaningless. And compare it with the company’s growth, profitability and balance sheet strength, because a higher multiple may be justified by superior returns and a lower one may reflect real problems.

This ties closely to the idea of margin of safety, which we covered in a separate article. A low multiple does not create a cushion if the earnings behind it are about to fall, or if the balance sheet is fragile. The ratio is only as reliable as the quality of the number underneath it.

A practical way to use the ratios

Begin by identifying what kind of business you are looking at and what drives its value, then choose the primary ratio that fits it. Check the quality of the numbers behind that ratio, looking for one-off items, accounting adjustments and the strength of the balance sheet. Compare the result with the company’s own history and its closest peers, and then bring in growth, return on equity or capital, and debt to see whether the multiple is explained by the fundamentals.

Finally, treat the ratio as one input into a wider view that includes business quality, management, governance and risks, and keep a margin of safety between the price you pay and your conservative estimate of value. No ratio can replace that judgement.

Follow our WhatsApp channel