What went wrong can be swiftly told. It was, and is, illogical to have a settlements system, Talisman, which is not connected to the dealing system, Seaq - but that was because Talisman predated Seaq, for it was designed for the old word-of- mouth trading on the Stock Exchange floor.
It is also illogical in this electronic age to have share certificates, which can get lost and have to be shuffled around the country. And it is rather primitive to have a settlements system that operates on a 14- or sometimes 21-day cycle.
So Taurus was to be the perfect, all- electronic, paperless system that would not just replace the British settlements system but also would connect into other settlements systems around the world for international securities. Of course, as with so many other grand British technical visions, they could not get it to work. Not only was it too complicated: it required legal changes and regulatory changes. And importantly, many of the personal customers for whom it was supposed to be a great advance did not want it: they rather liked the idea of having share certificates to show what they had bought.
The Bank of England now has to step in where the Stock Exchange has failed. Its task force will not report for three months, but the general way forward is very easy to see, for there is a model in the gilts settlement system. Equity settlements have to be split into two, retail and wholesale, with some sort of cut-off in terms of size of bargain.
For retail investors, comprising the vast majority of individuals buying and selling on the exchanges, the existing Talisman system will have to be upgraded. There is no reason why this should not be done, keeping the paper but using technology to handle it more efficiently. There is also no fundamental reason why Talisman should not move to a rolling settlement basis, say five days after the actual transaction. This would not be ideal, but it would cut the period in which a counterparty might default, the main weakness of the present system of two- and three-week accounts.
For wholesale investors do need a book entry system, which gets rid of the paper. Here the central gilts office, which acts as a clearing house for institutional gilt trades, is the appropriate model. It would be perfectly possible to set up a central equity office, which would act as depository for the shares in which the institutions trade, on lines broadly similar to that which operates in New York.
True, equities are more complicated than gilts. They have several registrars rather than a single one; dividend payment dates are irregular; and there is no concern about stake-building, so there is less need to be able to look through nominee accounts to see the ultimate beneficial owner of any particular block of shares. But it would be infinitely easier to build a system like this, which really is only applying the experience developed elsewhere, than it is to invent an entirely new system like Taurus.
So that is that. What are the other lessons? There are at least three. The first is that it is usually wrong to try to pioneer technology. It would have been far better to buy 'best available plain vanilla' from somewhere else, rather than try to develop a world-beater and fail. When the Stock Exchange changed its dealing system at the time of Big Bang, it bought in the Nasdaq screen system from New York and adapted it into Seaq. Since then, Seaq and its cousin, Seaq International, have worked very well.
Next, while having a decent settlements system does matter, it is possible to do very well with a less-than-perfect one. International equity trading is immensely competitive, yet London takes some 70 per cent of the world market (and 90 per cent of the European market) in cross-border equity trading. Over the past seven or eight years it has almost certainly increased its market share.
A lot of this success must be down to Seaq International, which has bitten into the still-considerable over-the-counter market in international equity trading. There may have been some slippage in recent months, but there is no clear indication that this is for settlements reasons. Of course there is a danger from greater Continental competition, for this is a business that Frankfurt and Paris would love to have. Because London's share is so large, it is not reasonable to expect some modest decline in the coming years. But in securities trading the main comparative advantage is in the breadth and depth of the market, not in slightly quicker settlements or slightly cheaper ones.
Third, the Stock Exchange has not been listening to its customers. It has many different types of customer: the member firms, the companies whose shares it trades, the institutional investors, the personal investors. The members were worried about the costs of Taurus, the companies were unhappy (and some had refused to help), the institutions were at best indifferent or at worst hostile, and any small investors who knew about it were disturbed about the additional charges they seemed likely to have to pay. It takes a particular arrogance to press on with something against this sort of resistance.
To be sure, Peter Rawlins, the chief executive, has now fallen on his sword, having narrowly survived earlier attacks. But that does not really solve the problem of an institution trying to adapt to a world that is radically different from that of 10 years ago.
Listen to this: 'I am confident that the necessary structure is now in place to achieve the successful development and implementation of the new settlement services. When introduced, these services will meet the requirements of the financial community and complement the attraction of our trading markets.' That was from the foreword to Project Taurus - Strategic Background and Cost Savings, a report of the International Stock Exchange, May 1990, signed by Andrew Hugh Smith, the chairman.
Dumping Taurus ought to be a blessing in diguise and technically it will be. But it also should lead to cultural change in the organisation, something that is much harder to achieve.Reuse content