Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ You Can't Get There From Here ]   |  Cooking the Books  |   [ It Beats Flipping Burgers ] >>

Weekly Column

Cooking the Books: How Clever Accounting Techniques are Used to Make Internet Millionaires

Status: [CLOSED]
By Robert X. Cringely

The late Frank Gaudette was Microsoft's first-ever Chief Financial Officer. He was also Microsoft's first head of Human Resources, first head of Facilities, first at running just about every department that had to do with operations but not product development, sales, or marketing. The wisecracking New Yorker was for a long time Microsoft's oldest employee, too, and contrasted sharply with the Generation X nature of the rest of the company. Gaudette came to Microsoft from Frito-Lay in the early 1980s and literally built the Microsoft we know today. And one day in the early 1990s, as we stood together at Microsoft waiting for an elevator to arrive, Gaudette shared with me the secret of knowing when to sell Microsoft stock.

My question was based on the idea that nothing goes up forever and there must come a time when even Microsoft is no longer a good buy. How can we tell when that time has come? It's eight years since that elevator meeting, Frank Gaudette is no longer alive, and Microsoft continues to be a phenomenal performer, but his advice still applies. Gaudette said to watch Microsoft's accounting practices. He explained that Microsoft carried on its books no value at all for its software. Assets like Microsoft Windows or Microsoft Office, which might be given some book value and depreciated over time were carried on the books as valueless. This contrasted at the time with IBM, which valued its software assets at billions of dollars.

"Watch for any changes in our accounting," said Gaudette. "If I need to I can start, depreciating the software and maintain earnings growth for years on flat revenue. Watch for the accounting changes, wait for the next uptick in the stock price, and then sell."

There is no reason why any investor with a long-term view should sell Microsoft at this time. But Frank Gaudette's advice makes a good point. Accounting practices can vary from company to company, and there is much to be learned by taking a closer look at the books. This particular example probably says more about its competitors' practices than it says about Microsoft's. Computer and software company accounting isn't as creative as that of the motion picture industry, where sleight of hand makes it appear that movies almost never generate net profits, but there is still plenty going on.

In recent years, the database software companies, especially Oracle and Informix, have used novel accounting methods to define what is or isn't a sale. Oracle Systems used to pay commissions on "sales" that looked to the rest of us more like vague intentions to buy. Why would a company want to pay sales commissions on transactions where no revenue was yet received? To maintain earnings growth and keep the stock price rising, of course. Rising earnings enable cheap stock acquisitions of other companies, keeping high tech companies competitive and fueling real growth. But if the share price falters, so does this system of empire building.

Oracle and Informix have cleaned up their acts in recent years, but there remains in the telecommunications and data communications industries another clever use of accounting to build "sales." It's frightening to think of how much of the Internet, for example, has been built with this technique that is commonly used by companies like Cisco Systems, Lucent Technologies, and Northern Telecom. And in this case, it is all perfectly legal.

Say I decide to become an Internet Service Provider, offering e-mail, Web surfing, and other services to customers in my town. In order to become an ISP, I'll need a connection to one of the many Internet backbones and a room filled with modems, routers, and servers. There is intense competition among providers of Internet backbone services and among builders of network hardware and software. These companies not only want the sale, but they need the sale to keep their booming industry booming. Their goal is always to do whatever it takes to get the customer into a long-term exclusive relationship.

Long before I ever sign-up my first Internet customer, I'll play these vendors off against each other. And the eventual deal I'll cut will not only supply my new ISP, it will for the most part finance it, too, through a system of hidden rebates, credits, incentives, delayed balloon payments, and options on equity investments.

We'll start with a contract for, say $10 million worth of equipment. This transaction can appear as either a purchase or a lease. Either way, the contract includes equipment, maintenance, support, and installation by equipment vendor personnel. The contract requires from me a 10 percent deposit ($1 million) and payment of the remaining $9 million over three years at about $300,000 per month. Remember, I still don't have any customers at this point, so I have no revenue. Fortunately, no revenues are yet required since the contract terms permit payments to be delayed, without penalty or interest, until the third year, when a balloon payment is due.

In this typical contract, the retail price of the equipment itself might be $7 million, with the rest of the sales amount allocated for installation, service and support. The net cost of goods is probably 30 percent of retail, so the supplier really has about $2.1 million tied-up in the equipment that's by now humming away in my machine room.

This is where things start to get complex. In either a separate contract or as part of the original sales agreement, I sell to the vendor options to buy up to 4.99 percent of my company. On a $10 million equipment sale, these options would typically carry a cash price of $1.5-$2 million. These options are convertible into Preference Class "A" Shares with certain conditions and terms that are impossible to change. By keeping the purchase to less than five percent of my company, there is no legal requirement to report the transaction. Not even analysts, much less shareholders of the vendor or the Securities and Exchange Commission need to know the deal has even happened!

A price of $2 million for five percent of my company values the entire company at $40 million and I still don't have any customers. I also don't have the full $2 million, either, since the equipment vendor pays itself from this option price the $1 million deposit required on the original order. This gives me a room filled with equipment and $1 million that I probably ought to spend on finding a few customers. And the equipment vendor books a $10 million "sale."

When is a $40 million company not really a $40 million company? When I default on that balloon payment in year three. Of course, I have every intention of building a good business and making the payment on schedule. But if for some reason I can't come up with the money, the equipment vendor has what's called an option offset against non-payment. This offset is the right to take over my operation to ensure payment on an "a priori" basis — basically a fire sale of all my company assets to cover the $2.1 million cost basis of the equipment. Here's the kicker — the equipment is already depreciated over the minimum three year period permitted to the buyer. The net effect is that the equipment maker can reclaim the equipment at its fully depreciated value and resell it for a profit.

In short, the sales department at the equipment vendor sells equipment with a cost basis of $2.1 million while the vendor's finance department buys options and equity for about that same amount, making it possible for me to become an Internet tycoon. Most importantly for the equipment vendor, it is a bookable $10 million sale for the quarter the transaction is signed. And because these companies like to use an accounting technique called Deferred Realization from Sales, the actual purchase price is spread over two to six quarters in amounts determined arbitrarily by the vendor. This allows the companies to report slightly better earnings on a quarter-to-quarter basis to satisfy the market's need for earnings growth. Revenues increase, earnings per share increase, stock prices go up, and more cheap paper is available to grow the equipment vendor even faster by acquisition.

Incredibly, this technique is perfectly legal and isn't generally seen as cooking the books. Frank Gaudette would have loved it. With deals like this, it's no wonder Internet companies are sprouting like weeds. And it is good that Internet traffic is doubling every 100 days, because if it didn't (or if it stops), this scam stops working, too.

Comments from the Tribe

Status: [CLOSED] read all comments (0)