Comment: When giants turn to the glue-pot

So ICI is worth about pounds 600m more cut in two than it is together. Market capitalisations are, at best, a rough-and-ready guide to any company's value, but yesterday's surge in the ICI share price shows professional investors would like to see the group broken up.

As far as ICI is concerned, there is nothing particularly new in this plan - the idea that it should be split into high and low-tech elements was floated privately several years ago by Sir John Harvey Jones, its former chairman. He felt the group would be intellectually more coherent, as well as being more attractive to investors. The market response yesterday certainly demonstrates the latter point.

While the idea may not be wholly new, the impact on the way other companies view themselves will be considerable. It forces large, diversified companies to ask the basic question: what is the glue that holds this enterprise together?

In Britain and the US we have recently seen the end of the second big merger boom since the war. Doubtless, there will be another, perhaps starting around 2010. In the meantime, there will be a period when shareholders, suspicious of grand merger plans, will hunt, instead, for value. It will take at least a decade for the damage to shareholders of visions like Saatchi & Saatchi to fade from mem

ory, and the best part of another before new grand alliances seem credible. But investment bankers are not idle: if they are not putting things together, they will start to take them apart.

Investor returns

In a low-inflation environment, companies will be hard-put to produce higher overall returns to investors than can be obtained in bonds. Shareholder pressure will force companies to examine their structure, and, if a break-up is likely to enhance shareholder value, do just that.

In one sense, this process of the break- up has already started. Already in the past five years there have been a number of examples of companies splitting themselves. BAT sliced off its paper and retail divisions when prodded by Sir James Goldsmith; Bowater split its US and UK arms; Courtaulds hived off its textiles. In the coming weeks British Aerospace will have to produce a plan that gets rid of several less profitable divisions.

But these examples are all cases where the various divisions should never have been put together in the first place. There was, in the case of BAT, little obvious connection between running supermarkets and producing cigarettes. (To be unkind, there is little synergy between cigarettes and financial services: at some stage BAT will be further divided. Ditto BAe's defence interests and Rover cars.)

Some companies have no internal logic, but have been assembled by some strong personality who has squirreled away, collecting businesses. These fall to pieces quickly when the personality goes over the rail. The really interesting cases will be those where the present conventional wisdom holds that the company is a credible entity. To see where this might happen, one has to consider the three sorts of glue that hold companies together, and ask how strong they will continue: industrial competence; financial discipline and control; and investor preference.

Industrial competence is, perhaps, the strongest glue. The core of a company like Boeing, Daimler Benz or Marks & Spencer is a powerful competence in a particular 'industrial' skill. It may be making things (like aeroplanes or motor vehicles), or it may be developing and distributing things (like knickers and chicken Kievs). Peripheral bits of those three empires may well be chopped off, but the depth of strength in the core is such that it is hard to see these groups being pulled to pieces.

On the other hand, it would be easy to point to other large, successful companies where there are several competences, but no single powerful one: Philips, GEC, BP are all companies that will look radically different in 20 years.

This is not a comment on the way they are run, though the better the performance of a company the more likely it is to be glued together by its financial discipline. This is a less strong glue than industrial competence.

Companies particularly adept at managing financial and human resources can grow fast: a group like Hanson is perhaps the best example of this type of enterprise, for it has grown by acquisition, but is held together by financial discipline. It is well-managed.

Moving from oil

But being well-managed may not be enough if the market or the technology moves away. If, as is not impossible to conceive, the world moves away from the use of oil as a fuel for transport over the next 30 years, even the world's best-managed oil company, Shell, may find itself unable to survive in its present form.

Investor preference is perhaps the weakest glue, for investors are so fickle. There was, in the 1960s and 1970s, a phase where investors wanted conglomerates, and companies like LTV in the US were assembled to meet that investment need.

A similar investor preference for multi-functional service industries led to the growth of Saatchi & Saatchi in the 1980s. That group was not created out of thin air: it was possible to raise the money to make all the acquisitions because there was an investor demand for the product.

In the low-inflation 1990s, it looks increasingly as though the prime investor preference will be for companies that are closest in their financial characteristics to bonds. Investors will want a steady, reliable, if unspectacular, flow of dividends. If they can get capital growth, that will be gratefully accepted; but investors will not particularly expect it, and they will be suspicious of any such claims.

This will change company strategy. Expect Japanese companies to concentrate much more on short-term profit for shareholders rather than supposed long- term growth, or market share. But it will also change company structure.

Companies with divisions that deliver the steady, unspectacular growth of dividends will find themselves under investor pressure to focus on those. But someone will have to undertake the risky or cyclical, high-reward activities, too. They will therefore have to find alternative structures for any divisions falling into that category.

The return of the private company, with many more Virgin Atlantics? Or a boom in specialist high-risk investment trusts? Probably both.