Sean O'Grady: What can go down, can go down lots further

The Private Investor

To tell the truth I was always a bit uneasy about my holding in Railtrack. I mostly feel a degree of affection for most of the shares in my portfolio, sometimes misplaced of course, and dangerously sentimental. But Railtrack was always something I didn't feel very good about, for obvious reasons.

I bought it, not in the privatisation but earlier this year, post Hatfield and post Gerald Corbett, when it was looking "cheap" at about £4, when, surely I thought, the worst is behind it, and it would soon grow up to be a model corporate citizen. It died young instead. They say you should never invest money you can't afford to lose and you should always be prepared to lose the lot. But I cannot say I was expecting to lose the lot not once but twice in four days.

The successive liquidations of Atlantic Telecom and Railtrack have rendered your correspondent's holdings in these companies worthless. Yes, the Government effectively renat- ionalised Railtrack on the cheap and yes, the bondholders pulled the plug unnecessarily early on Atlantic Telecom, but really I can't complain about my losses in the way I would if, say, I had been mis-sold a pension or an endowment policy.

I have made my own mistakes. As Lady Bracknell might have said: to lose one company in one's portfolio may be regarded as a misfortune; to lose two looks like carelessness. Was I careless? Maybe. I certainly fell for a few fallacies about cheapness. I would not say there is no such thing as a cheap stock, but I will have to develop a much clearer idea of what constitutes value. With Atlantic, for example, I first bought at a little over £8 per share, when "it looked cheap" compared with its peak of about £12.

But that was when technology ruled the Earth and BT was at £15, "traditional" valuation methods were declared obsolescent and everything seemed possible. Except, that is, what actually happened. Atlantic dropped below £5, £2, £1, and all the way down below 10p. They reached a nadir a fortnight ago at 2p and were standing at 5p when the stock exchange suspended them. They now rest at the most psychologically important level of all – sod all.

I thought Atlantic looked cheaper and cheaper compared to its previous levels. So I added to my holdings, bringing the average cost down, until Atlantic was basically in penny share territory and it was hardly worth selling it. I had long since skipped the opportunity of baling out, and maybe there was a degree of vanity there, and I like to "buy and hold".

The chart shows the dramatic rise and fall of this tiny Scottish telecoms outfit. Someone out there, probably, bought at an even higher price than I did and has lost more. But that is no comfort. It is a painfully embarrassing mistake but one has to examine one's own errors of judgement in the harshest of lights, that is, if one is to extract maximum educative value from this expensive lesson.

That harsh light exposes a number of fallacies. First, that if a share has fallen by 70 or 80 or 90 or 99 per cent of its value it can't fall much further. It can, of course, still fall by 100 per cent. You know that too, but until it happens to you, you still never quite believe it.

Second, that "psychological" so-called barriers and levels don't really mean very much, even though they assume such overweening importance when people write and think about investment. One of my earliest "economic memories" was the Sterling crisis of 1976, when the then Labour government and in particular that Chancellor of the Exchequer, Denis Healey, were battling to keep the pound from slipping below the politically significant $1.50 level.

If the pound managed to finish the day at $1.5001 Mr Healey was a national hero; when it went to $1.4999 Denis became a symbol of our painful national decline. But it didn't affect the price of a Datsun Bluebird in the showroom, or indeed, the long run value of sterling. It was just a Summer of '76 game, like guessing when the great drought would end. I should have remembered that.

Round numbers don't matter, with the possible exception of the ephemeral effect of people buying a stock simply because it has fallen below what they fondly imagine is a "support level". As every former Atlantic shareholder will attest, there is no such thing.

s.o'grady@easynet.co.uk