New Math for a New Economy

Remember: This system was invented hundreds of years ago. Luca Pacioli, the monk who created it, was a genius. He developed a system that is still working 500 years later. But Pacioli's system is frail. After all, it relies on transactions. But when you're working with knowledge assets, value is created or destroyed without making any transaction at all.

When a drug passes its clinical tests, huge value is created -- but there's no transaction. Nothing changes hands. Nobody buys anything, and nobody sells anything. When software passes a beta-test, it suddenly becomes valuable -- but there's no transaction. Or think about how value is destroyed: When a big, old company is late in figuring out how to enter the world of e-commerce, huge value is destroyed -- but there's no transaction.

If something as fundamental as the accounting and financial-reporting system doesn't work in the new economy, why hasn't there been a more vocal call for change?

There are two major barriers to change. The first is an objective difficulty to this problem: The issue of knowledge assets is inherently uncertain -- we're still struggling to come up with a definition that works. And the issue of intellectual-property rights, for example, continues to be fuzzy. There's no one solution that will eradicate the problem -- and every new solution presents new problems. Given how difficult the material is, it's easier for people to wait for a better solution to come along.

The second barrier to change is an informal coalition that opposes any change to the current system. Managers love the current system. They don't want to put anything on the balance sheet that may turn out to be worthless. Accountants share this love of the current system. If they don't have to value intangible assets, such as AOL's customer-acquisition costs, their legal liability is reduced. Let's face it: Valuing things that are inherently difficult to value, and then standing by that valuation when someone sues you, can be very unpleasant.

Institutional investors and financial analysts are also quite happy with the current system because they think that they've got inside networks and proprietary information. They have lunch with managers. They visit companies. In doing so, they feel that they're getting important private information. How would it serve their interests if that information were made public? So there are some awesome forces against change.

If all of these people are against change, then the system must be fine, right? What's the problem?

The problem is that lots of mistakes are being made. For example, I just finished studying roughly 1,500 companies -- all of which have significant R&D investments. About a quarter of these companies are systematically undervalued by their investors. And many of them are computer, biotech, and software companies with substantial R&D but below-average earnings. That means that the cost of capital for these companies is unusually high -- and that impedes their growth. The systematic undervaluing of these companies brings serious economic and social costs to the companies, to their shareholders, and to the economy. So the problem is not academic or abstract; it has serious business implications. And although I do not have the data to prove it, my guess is that many internal decisions are also deficient because managers, too, are relying much too heavily on accounting information.

What solutions do you propose to fix that problem?

I think that there are a few remedies -- none are complete, but all will help. One is to improve the accounting system. Some people have given up on accounting altogether. They say that the system is dead and that something entirely new is needed. To me, that would be a big mistake. I believe that accounting is still incredibly efficient. So the solution is not to do away with the old system but to improve it.

I don't expect any breakthroughs but slight amendments that improve on what already exists. One example: satellite accounts. These can be a set of accounts around the regular ones that will provide more information about the real value of assets. The U.S. government, for example, expenses R&D in the same mindless way that companies do. But the government has also set up satellite accounts in which it capitalizes R&D. It's not a great revolution, but it does provide a way to compare things.

And your more revolutionary proposal?

Another remedy requires going outside of the existing system. I've developed a way to measure knowledge assets, intellectual earnings, and knowledge earnings. It's a computation that starts with what I call "normalized earnings" -- a measure that's based on past and future earnings. When you're dealing with accounting for knowledge, you simply cannot do it unless you consider the potential for future earnings that knowledge creates. In fact, that's one of the things that is fundamentally wrong with all of the other ways we have of accounting for earnings, including improvements such as EVA [Economic Value Added]: They are all based purely on history. They are accounting in the past.