Infosys

When the weighing machine disagrees with the crowd

Graham taught that in the short run the market is a voting machine, registering popularity, but in the long run it is a weighing machine, registering worth. Few situations show the gap between the two as plainly as Infosys does now. The votes have gone against it — the shares are down about thirty per cent in a year and rest near their lowest in twelve months — yet the scale, when you actually weigh the business, reads heavy.

What does the business do? Infosys writes, runs and modernises the software that large companies abroad depend on — banks, insurers, retailers, manufacturers across America and Europe. They hand over the work; Infosys does it and bills for it, year upon year. It is India's second-largest firm of its kind, valued at about four lakh twenty thousand crore rupees, and the engine it runs is among the most profitable in the land: roughly thirty-two per cent earned on shareholders' equity, forty on capital, and not a rupee of debt to service.

A first-class business at a third-class multiple

The heart of the matter is the price. Infosys changes hands at about fourteen times its earnings. Weigh that against its own past: the median multiple over the last ten years has been close to twenty-four, and over five years close to twenty-seven. The market is asking the patient buyer to pay barely three-fifths of what it has itself, on average, thought the company worth across a decade. That is a margin of safety carved not from a troubled small-cap but from a national blue chip — the discount sits in a fortress, not a shack.

The defensive investor's checklist, which Graham insisted upon, is largely satisfied here. Earnings have grown without interruption for years — profit compounding at roughly nine per cent over five years and accelerating to the low teens of late — and the dividend record is long and generous, a payout of about two-thirds of profit that hands the owner a yield near four and a half per cent. On the question of dilution, the answer is the best kind: there has been none. The share count has, if anything, *fallen* as the company bought its own stock back — equity capital edged down from about ₹2,098 crore to ₹2,024 crore over four years — so the owner's claim has quietly grown rather than been watered. Sales, meanwhile, have compounded near eleven to twelve per cent over the longer haul. This is not a melting ice cube bought for its drip; it is a sound, growing enterprise temporarily out of favour.

Where the danger lies

A careful man weighs the risks before the rewards. The reason for the cheapness is genuine: near-term growth has cooled to single digits — about seven per cent over three years — as Western clients delayed their spending, and foreign institutions, who held over a third of the company, have pared back toward twenty-eight per cent, which is the selling that drove the price down. The shares are not cheap against book value — over four times it — so the safety rests on earnings power and the discount to its own history, not on tangible assets one could liquidate. The promoter stake is low, near fourteen per cent, the ordinary condition of a founder-built giant now run by professionals. Should global technology budgets stay frozen for longer than expected, a cheap share can remain cheap for a while.

What this asks of you

This is a roughly one-year holding for the temperament Graham prized — the willingness to buy a thoroughly sound business precisely because the crowd has lost interest in it. The protection is threefold: a multiple well below the company's own decade-long norm, a fat dividend paid while you wait, and a balance sheet with cash and no borrowings. The thing being bought is quality on sale, with the margin of safety supplied by the gap between price and the firm's long-established worth. Buy it for that gap and that dividend, hold while the weighing machine does its slow work, and let the eventual return of patience to the sector — not any heroic forecast — be what pays you.