Beauty is in the eyes of bidder

What are companies really worth? Like anything else they are worth what people are prepared to pay for them. Yesterday's stock market illustrated the clash between what a company is worth to a competitor and what it is worth to an investor. Start with the share price movements of two companies in the same industry, Wellcome and Zeneca. Wellcome naturally shot up in response to the Glaxo bid which at 1025p priced it at some £9bn. This price was one-and-a-half times the level of Wellcome shares on Friday. So the value of this company to Glaxo was over £4bn more than the value assessed by the mass of shareholders.

This always happens in takeover bids: such bids can only occur if someone believes the market is undervaluing the company concerned. But yesterday also pushed home the opposite lesson: that there are equally times when the market puts a much higher valueon part of a firm that the current owner. Remember that Zeneca, the shares of which also rose though less markedly to price it at about £9bn, was once merely the pharmaceutical division of ICI.

Contrast that with the price tag of £5.5bn on its former parent. ICI sold it because the market put exciting pharmaceutical companies on a much higher rating than boring chemical ones.

So there we are, two giant companies in the same industry, both worth about £9bn. Yet apparently it is logical for one to lose its identity by being merged with an even larger one, and for the other to have gained a separate identity by being demerged from what was at the time an even larger one.

There are two common ways of commenting on the process of merger and demerger. One is to start from the premise that markets are deeply capricious and accordingly there are opportunities for companies to be acquired at much less than their true value.

The other is to talk in the language of the business school - in terms of industrial strategy, management skills, core competences and so on.

The rationale behind the Wellcome takeover, looked at this way, would be the need for the pharmaceutical industry to consolidate; the case for Zeneca to be valued so highly as a separate entity is the lack of synergy between a lowish tech chemical industry and a high-tech drugs one. Both approaches are perfectly valid in their own way. The markets are absurdly capricious in their valuations, and their absurdity does create opportunities for takeovers and demergers which might not be justified on any other grounds. And there is such a thing as industrial or managerial logic - sometimes assets are genuinely worth more to one company than they might be to anyone else because it is able to make better use of assets.

But the very fluidity of company structures nowadays - the fact that, like Velcro, they can be stuck together or pulled apart at will - raises the crucial question: What is the glue which holds firms together?

Look at it this way. Large companies can, and do, manufacture wherever they like in the world. They can raise finance on any capital market and on much the same terms - indeed most seek to have a quote on several exchanges so that they can attract an international shareholder base. They can locate their research and development wherever the skills are best located. They can, increasingly, move their headquarters to the place which offers the most attractive tax status. And they can even hire their top management from an international talent pool, though for the moment the concept of the international management star has yet to take hold. The international nature of the market for business talent is often cited as a justification for high salaries, but in practice there is still only limited cross-border trade in senior executives.

Add to all this the fact that corporate structure has become completely fluid and it becomes hard to discern the rules of engagement. It was easy to value companies when they were largely national entities, using large fixed assets to manufacture a product or deliver a service. That is all history now. What, in this shifting world, gives a company its value?

There is no simple answer to that, but I think the Glaxo/Wellcome and the ICI/Zeneca stories do help clarify the issues: put crudely, one is driven by the logic of the industry, the other by the logic of the markets.

The position of the industry is straightforward. After two decades of very rapid growth the pharmaceutical industry is caught in a vice.

On the one hand the costs and risks of developing new products requires enormous size. On the other, the squeeze on margins from mass purchasers (managed heath-care schemes in the US, the squeeze on public sector funding in Europe) puts the whole industry under great pressure.

In all industries which need to stay big, but have to cut costs, the almost inevitable result is a spate of mergers. The Glaxo/Wellcome proposal is a classic response. ICI/Zeneca split, on the other hand, was crafted for market preferences.

Institutional investors need their products in neat categories, so they can walk down the supermarket shelf, shopping in an orderly way - a bit of exposure to a cyclical industry, a chunk of a growth company and so on. Zeneca is a "pure" product, doing one thing. Whether or not, at £9bn, it is big enough remains to be seen. One of the reasons for the rise in its shares yesterday was its attractions as another bid candidate.

This clash of values seems bound to continue. Takeovers and mergers go in waves and we are in the early stages of a new round of these. They will occur most frequently in industries where the divergence of perception is most stark.

And the glue? Will it be industrial logic or market preference? In reality, it must be both. But in applying the twin disciplines remember that some companies work very well despite fitting either pattern.

Imagine a company which runs an airline, sells soft drinks and financial services. Imagine another which makes high-tech defence kit, cars, and fashion magazines. Nuts? No, both are very successful; the first is British and is called Virgin; the second is French and is called Matra. But both are privately controlled. Maybe the markets need a new category of companies: good ones which don't fit.