A vital reality check for would-be investors


Can Google, founded less than a decade ago, really be worth $35 billion? A short guide to the basic economics of asset and business valuation leads to a clear answer.

Could Google, founded less than a decade ago, really be worth $35bn? In a competitive market, the maximum value of an asset is the cost of replicating it. The economic principles could hardly be simpler. If it is possible and profitable for companies to do something, they will. A competitive market is one in which others are free to do anything that they see is rewarding for someone else.

The replication principle does not help much in valuing Barclays Bank, General Electric, BP or Microsoft. Barclays Bank is one of the few survivors of the hundreds of small banks that existed in Britain two centuries ago. Its family tree is complex and impossible to reproduce. You could establish new businesses such as those you find within General Electric, although it would be difficult to create a new manufacturer of aero engines from scratch. You would certainly find it a long and costly process to reproduce the depth of management talent in that corporation.

An integrated oil company such as BP is like an intricate machine: even if you put together all the component pieces, it is unlikely that they would fit together with the precision achieved when the parts have meshed with each other over a long time. The complexity of most large, well-established businesses makes them hard to replicate. And you cannot replicate Microsoft either because the law protects the copyright in its software. All of these businesses are unique and the only way to measure their worth is to forecast their prospective earnings far into the future.

But the replication principle is very helpful in appraising new businesses in markets where entry is easy and frequent. A mobile phone company is worth the cost of building its network and of acquiring its customers. A telephone engineer can estimate the first and a marketing consultant the second. The sum of these two is much less than the market valuation. If you had made such a calculation, you would not invest in mobile phone companies. The replication principle is not an alternative to discounted cash flow calculations and earnings projections, but a check on them. It tells you what is likely to happen to cash flow and earnings as competition intensifies and the market matures.

Analysts often try to escape this reality check by talking of the value of the brand. Sometimes they are right. The Disney Corporation is worth a lot, for the legal reason that you are not permitted to reproduce its characters, and the practical reason that not many people share Walt Disney’s genius. Coca-Cola is the world’s most powerful brand because of a hundred years of history, a secret formula and an unparalleled distribution system.

Leading universities, legal firms, even companies producing soap powders, require a lot of time and investment to replicate because consumers have learnt to trust the quality of their product. These brands derive their value because people will prefer them to functionally equivalent products. There is a key commercial distinction between the name of an excellent product and a brand that transcends the underlying product attributes. The by-line on this column becomes a brand only when my name prompts you to read on rather than to turn the page, or to believe something you would not have credited from a less illustrious columnist. It is rare that a new writer, or a new business, will command this brand value. The replication principle is at work again. The value of an asset is capped by the cost of reproducing it, and what was recently created is usually more easily reproduced.

Imitation takes time and innovators can earn substantial rewards in the transitional period before others understand why their idea was such a good one. Even in high technology markets, where demand is often focused on the best product available at the time, innovation produces a flow of new entrants – look at the history of spreadsheets, where Lotus overtook Visicalc and Excel overtook Lotus. But the better the idea, the shorter that interval is likely to be.

This column is not in the business of giving investment advice. But you would be correct to infer that Google will remain on my favourites list, but not in my portfolio. You can do a lot of software engineering for $35bn.

Print Friendly, PDF & Email


Please enter your comment!
Please enter your name here