Companies that buy in their own shares are, as the saying goes, paying some shareholders to go away at the expense of those who remain. However, there is always the exception that tests the rule, and last week Next again showed that it is an exceptional company. In making clothes retailing look easy, it also generates more cash than it can profitably invest.

The textbooks say the excess should go back to the shareholders, either in higher dividends or in return of capital, but Next is, philosophically at least, a family business, and dividends mean a tax bill, while it quite suits Simon Wolfson, the family scion in charge, to keep the capital where he can see it.

The result is a steady, determined buyback programme. Over the last 12 months, Next shares have averaged around £32, and the company has bought in 10m shares, or 6 per cent of the equity. With the price now above £38 after another pleasant trading surprise, those remaining have profited at the expense of those who sold.

Now look at GlaxoSmithKline. It seems that there are so few attractive research or takeover prospects that the board can think of nowhere better to invest the company’s capital. In the last 12 months GSK has bought in 150m shares, or 3per cent of the equity. Despite this support, the price has fallen from £14.80 to £13.60.

GSK joins a long list of big companies which have bought in their own shares at prices which later prove to have been inflated. Worse still, some boards suspend their buyback programmes when share prices fall, suddenly fretting about the balance sheet. Next’s example shows that buybacks can boost shareholder value. It also shows why, in most cases, they don’t.

Take a long, long view

Once upon a time, there were 50 companies which stood apart from the rest. Investors who wanted to sleep easily bought their shares, inevitably nicknamed the Nifty Fifty, companies so strong and powerful that their shares commanded 30 or even 40 times earnings. The rating of the 50 reached its peak in 1972, when shares in Coca-Cola and Xerox each cost 46 years’ earnings.

Over the following 25 years, the ratings fell to earth, while the actual performance of the companies allowed Jeremy Siegel to work out whether those eye-watering ratings had ever been justified. Some had turned out to be absolute dogs. Xerox’s performance warranted a P/E of 18 rather than 46, but Coca-Cola had continued to rot the world’s teeth so successfully that it would have been worth paying 92 times 1972 earnings.

There is no equivalent to the Nifty Fifty today, but there’s a groundswell suggesting that those massive, defensive international stocks which might be in a latter-day version of it look expensive after last year’s gains. Siegel’s analysis suggests that the better ones are not. Of course there were better buying opportunities for the Nifty Fifty after 1972. There almost always are, although you only get in at the bottom by luck. Patience, conviction and eternal vigilance are more helpful than attempting market timing.

High ratings for big companies can be justified. Philip Morris’s profits would have warranted a P/E of 78, rather than the 24 the shares commanded in 1972, despite the decades of government hostility. Siegel’s work also contains a warning about the ephemeral nature of technology stocks; Burroughs, too, was considered to be worth 46 times earnings back then. Subsequent performance (it merged to become Unisys in 1986) showed that it was really worth less than a tenth of its rating. Given that history, and the more recent experience with Nokia, it’s little wonder Apple’s P/E looks absurdly low today.

Get your gong up front

We may never know quite why we must now call him Sir Hector Sants, but we can take an educated guess. The golden rule for gong-seekers is to get your demand in at the start. Phrases like “we’ll take care of you in due course, old boy” mean that they won’t. It’s only when the politicians are suitably desperate that they will really commit. Sir Hector may not be an obvious choice for honours in the aftermath of the financial crisis, but he resigned in dudgeon following the last election. Only the discrete charm of the new Chancellor persuaded him to stay on…

This is an updated version of my column in Saturday’s FT