Creative Accounting

I HAVE BEEN reading with some fascination this week the Washington Post's four-part investigation of MicroStrategy , one of those rags-to- riches-to-rags stories of the dot-com bust.

Coupled with the announcement that the US Department of Justice is taking over from the Securities and Exchange Commission and mounting a criminal investigation into the recent Enron crash and last month's examination in Business Week of current practices in reporting earnings ( here), it's highlighting the fact that although investors have far greater access to information about the companies they put money in, it's increasingly difficult to wade through all that information and arrive at any realistic assessment. I'd say it was a job for professionals except that MicroStrategy's and Enron's accounts were all signed off by their auditors, people like PriceWaterhouseCoopers and Arthur Andersen (whose employees, according to this morning's Independent, destroyed a few relevant documents).

Accounting standards are now confusing enough that it seems likely we'll be seeing a few more of these scandals over the coming months. Are earnings are generally accepted accounting principles (GAAP), operating earnings, or the pro- forma earnings beloved of high-tech companies, especially those who can turn losses into profits by excluding lots of expenses? Are restructurings, layoffs, and factory closures special charges or the ordinary cost of doing business? And should companies be required or allowed to include investment income and losses as part of their ordinary business?

The boom years saw a number of companies set up internal venture capital funds to invest in start-ups. Intel Capital, for example, has investments in more than 550 companies in 20 countries ( here). During the boom years, a number of companies, Intel among them, benefited handsomely from these funds. Even smaller companies did well out of the stock market during the boom when they spun off subsidiaries. RSA Data Security, for example, got a nice boost to its earnings for several years from selling off portions of its stake in its former subsidiary, Verisign. It's not the company's fault that online tables showing company data did not distinguish between the one-time proceeds of those sales and its actual business earnings in calculating the company's price/earnings ratio.

Business Week blames the dot-com boom. It's certainly true that during the most, er, optimistic period classifying all kinds of charges - marketing expenses, say - as "special" allowed a number of companies to inflate the speed of their success. Other questionable practices of the time ( here) included booking barter deals - you carry my ads, I'll carry yours - as revenues, booking the full sale price of third-party products instead of just the business's commissions on them, booking rebates as marketing expenses, capitalising marketing costs, and so on. Capitalising marketing costs was the issue between the SEC and AOL ( here) when AOL restated its earnings for 1995 and 1996 under new rules, the accounting change wiped out all the profits the company had ever made.

That example shows how non-trivial these niceties can be. Business Week highlighted the difference by comparing the average earnings per share for the S&P500 under GAAP ($6.37, for an overall P/E ratio of 38), S&P itself ($9.17), and the figures from First Call calculated from analysts ($10.78, for a P/E ratio of 23). There's also the problem that sometime in the last few years people have shifted from valuations based on last year's actual earnings and this year's "projected earnings".

It's the equivalent of my saying to the bank when applying for a mortgage, "Hey, I know I only earned $20,000 last year, but that's OK, I'm figuring on earning $50,000 this year, so my request for a $150,000 mortgage really isn't out of line."

Related to this is another common problem involving when revenue is recognised. If you, an individual, get a contract for work that will pay you $50,000 a year for ten years, you don't claim the whole $500,000 as revenue in the first year. Some companies do. Nor do you claim revenue when you sign the contract; you claim it when you get paid. But some companies do the former; in May 2001 the Motley Fool pointed out that a tiny note in RSA's most recent 10-Q annual filing indicated that the company was switching to recognizing revenues when products were shipped to distributors. ( Here).

But of course anyone who thinks these things started with the dot-com boom is wholly out of line. You have only to read the great classic The Intelligent Investor by Benjamin Graham to see that companies have always tried to get away with as much as they can in making their quarterly reports look favorable. Graham also highlights something we can expect to see a lot of this year: the big bath. That is, one side effect of the September 11 attacks will be that companies will take advantage of the public's hugely lowered expectations for the year to wash out anything and everything they can. Then they'll all look so much better in 2002.

Last time I wrote about this, in early 2000, the SEC was proposing to crack down on such practices. They are, as recent events have made clear, going to have to do a lot more work. µ

Previous Columns
Dumber people can run Windows
2001 in review
Care in the community
Remembran ce of postings past
BT's Stupid Patent Tricks
Preserving our freedoms
It's beginning to look a lot like Christmas
Net is the mother of re-invention
Save the Cookie
Digital rights and the new era of world terrorism

Wendy M. Grossman, whose Web site is pelicancross ing.net, is author of From Anarchy to Power: the Net Comes of Age (NYU Press, 2001), net.wars (NYU Press, 1998), and the Daily Telegraph A-Z Guide to the Internet (Macmillan, 2001). She can be reached at this email address.

Copyright on all articles published in the INQUIRER is hers.