Calculating accurate values for companies in the new economy takes more than a grasp of mathematics, says John Kay

Do the math. The slogan favoured by Jim Clark, creator of Silicon Graphics, Net- scape and Healtheon, has become the mantra of a generation of consultants and investment bankers. The new economy, they claim, requires new principles of valuation.

C.com is one of the most exciting prospects in business-to-business commerce. It is the world leader in a growing market - annual sales by 2010 are likely to be $500,000bn. If C.com can maintain a 5 per cent share and earn only 1 per cent net margin its prospective annual earnings will be $250bn.
If we assume that market growth after 2010 is 5 per cent and discount future revenues at 10 per cent, the prospective value of C.com is $5,000bn - about 10 times the recent market capitalisation of Microsoft, Cisco or General Electric.

You don't have to wait for the IPO. You can buy shares in C.com right now for less than 5 per cent of that value. C.com is called Citigroup and in addition to its foreign exchange trading, which is the business I have described, you get its other wholesale, retail and investment banking activities and a leading insurance company thrown in.
Of course, nobody would be so stupid as to value Citigroup in this way. Yet I have followed more or less exactly the methodology recommended in the latest McKinsey quarterly for the valuation of new era companies. They use precisely analogous calculations to arrive at a valuation of $37bn for Amazon.com.

Paul Gibbs, head of merger and acquisition research at J.P. Morgan, recently used similar principles to confirm that assessment of Amazon. He then performed the same calculation for internet service provider Freeserve.
Assume that UK retail sales grow at 5 per cent a year, that 25 per cent of sales take place on the net, that portals capture 50 per cent of these, that Freeserve gets 30 per cent of the portal share and maintains an 8 per cent commission on sales. Multiply these together and you establish that in 2017 Freeserve will make profits of £2bn ($3.2bn). This, he argues, justifies a value today of £6.50 per share.

But I prefer T.com to Freeserve. T.com has a customer base four times larger than Freeserve. Its franchise is stronger. Its customers are concentrated in the affluent south-east of England, where it faces virtually no competition. Market research shows that more than 95 per cent of its customers use its essential services every day, many of them several times a day.
T.com also has ambitious plans for expansion. At present, its geographical coverage is less than one-quarter of the market in England and Wales. T.com has an interest in south-east Asia. The population of China is 100 times the number of people who today can access T.com services. Even after the recent market correction, T.com, better known as Thames Water, still has a market valuation below that of Freeserve. At the widest point between the old and new economies, Freeserve was worth four times as much.

The reason the Citibank calculation is nonsense is simple, but fundamental. The margins Citibank makes on its forex business vary widely. If you buy small quantities of notes from a bank, the spread is much wider than 1 per cent. If you are a large corporation trading major currencies, the margin is wafer thin. Entry and competition force prices down to the related costs.

In Mr Gibbs's model, Freeserve earns profits of £2bn, about equal to the current profits of Tesco, J. Sainsbury and Marks and Spencer together. And it earns these on revenues of only £2.5bn, so that profits are 80 per cent of the value of its sales. No established business earns margins of that size.
Thames Water is one of a tiny number of companies whose market position is so strong, whose output is so necessary to life, that if it charged five times the current price we would have little choice but to pay. We do have one option - to insist the government intervenes. It confines Thames Water to a return on its capital employed of about 6 per cent.

There is a key formula in the new math. The required yield on a security is equal to the difference between the rate of return demanded from that class of securities and its expected rate of growth. So, if you expect a return of 5.5 per cent from a share whose dividends will grow at 5 per cent, calculations show that the dividend yield should be 0.5 per cent.

This is the calculation done by James Glassman and Kevin Hassett in their book Dow 36,000*. Claiming that sustainable dividends average half of earnings, this yield equates to a price/earnings ratio of 100 - implying a target of 36,000 for the US index.

The trouble with this theory is that it takes too long to produce what the investor is looking for. Investors will receive only two-fifths of the cash sustaining the valuation this century. One-third of the total depends on dividend cheques that will arrive after 2200.

Two hundred years ago, well before the Dow Jones average, prudent, diversified investors would have owned slave traders and sugar plantations, or perhaps a bold speculation in that symbol of the then new economy - a canal. The next 200 years may be more stable than the last and Microsoft and Cisco may prove more enduring than plantations. But we can hardly be sure.

While 5 per cent may be a reasonable assumption for the growth rate of dividends in the US economy as a whole, it is likely that well before 2200 most of these will come from companies not yet founded.

Suppose we accept Glassman and Hassett's protests that their dividend growth expectations are conservative. If you raise them only from 5 per cent to 5.25 per cent, the anticipated value of the Dow Jones average is 72,000. At 5.5 per cent the formula breaks down because the shares of US companies are infinitely valuable. Not even the most credulous dotcom investor believes that.
The math also works in reverse. If expectations of return are 6 per cent rather than 5.5 per cent, the market p/e ratio falls to 50 and the value of the Dow Jones from 36,000 to 18,000. And if equity investors require a return of 8 per cent, you would have to conclude that shares are worth only half their current level.

An 8 per cent expected total return is not ambitious for an equity investor. A day trader might think you were talking about a weekly profit rather than annual. Glassman and Hassett use a figure as low as 5.5 per cent - the return on Treasury bonds - because they argue that equities have so consistently outperformed bonds that there is now no risk associated with equity investment. In other words, because equities are sure to offer higher returns than bonds, the expected yield on equities should be the same as on bonds.

You do not have to be Wittgenstein to spot the flaw. The rules of logic hold even in cyberspace, and so do the principles of economics. Profits are hard to earn in competitive businesses, and markets that are not competitive are usually regulated. The value of companies ultimately depends on their capacity to generate cash for shareholders. Distant returns are uncertain. Share prices are volatile, and investors need to be compensated for the risks. These truths are as valid in the new economy as the old.

By all means do the math. Isaac Newton, who could do the math better than most, gave up an annuity of £650 per year to invest in the South Sea Bubble. In addition to the math, you need the econ, the pol, and perhaps the psychology.
* Dow 36,000: The New Strategy for Profiting from the Coming Rise in the Stock Market, James K. Glassman & Kevin A. Hassett, Random House, 1999.
 
 

Source: John Kay, FT,  April 23 2000
              The author a director of London Economics.