Hiding in plain sight

With Martha Stewarts's trial winding down, the Rigas family trial about to get underway and former Enron CEO Jeffrey Skilling's trial being an estimated two years away, it may be a good time to focus on the next big corporate scandal.    (Eliot Spitzer, if you're reading this, you might want to take notes.)

In my view, the next "scandal" will be the public's realization that using stock options to pay employees has helped companies (and particularly tech companies) systematically overstate earnings by one third or more for many, many years.    First a bit of background, for those of you who haven't been following the issue.

Companies have long issued options to senior executives giving them the right (but not the obligation) to purchase a fixed number of shares of their company's common stock at a fixed price (usually the market price when the options were issued).     In most cases, these options expire after ten years, vest over several years (usually five) and must be exercised before the employee leaves the company.    Since the ultimate value of the options grant depends upon the future price of the company's stock (which is unpredictable) and have no current monetary value when issued because the exercise price equals the market price, APB 25 allows companies to record no expense when issuing options.    However, stock option grants do have value (as anyone who as ever purchased an exchange traded option knows) and often make up a substantial portion of total compensation for key employees.   Tech companies have particularly liked using stock options to pay employees because not only are they not a drag on profits, but they also require no use of company cash.     (In fact, the exercise of employee stock options is a source of cash as new shares are issued and sold to the employee under the option agreement.)

Sophisticated investors have long been aware that stock option grants are not reflected in the income statements of US companies and that therefore reported earnings overstate profits for companies that make extensive use of options to pay their employees (including most tech firms).     In fact, footnote disclosure of this expense item and its impact on earnings has been required by FASB Statement 123 since 1996.    However, widespread awareness within the financial world has not inoculated brokerage or accounting firms from liability in past scandals like the savings and loan collapse (where accounting fictions like "regulatory capital" were included as assets on balance sheets) or the internet bubble. 

It is unlikely, however, that the average private investor is aware of how significant options accounting is for the earnings of tech companies.    I did a quick and dirty analysis of how much reported fully diluted net income would have been reduced had employee options issuance been expensed for the ten largest components of the Nasdaq 100 Index, which collectively accounted for more than 40% of the index and $1,034 billion in market capitalization.

As you can see, excluding employee options as a cost of doing business increased reported net income per share by an average of 77% in 2001, 61% in 2002 and 34% last year.    For some companies, like Cisco and Ebay, for example, earnings per share were doubled or tripled due to the accounting fiction that the cost of employee options grants should not be considered an ordinary cost of doing business.

While many traditional industrial companies have also benefitted from not expensing employee stock option issuance, the impact has not been as significant.    The table below shows the same analysis for the ten largest components of the Dow Jones Industrial Average, accounting for more than 50% of the index and $720 billion in market capitalization : 

For some of these blue chip firms, notably tech stalwart IBM, excluding employee options expenses increased earnings by a significant 33% in 2002.   But for most of these firms, options accounting increased EPS by less than 10%, and for some firms, like Wal-Mart or Altria, expensing options would have reduced net income per share by only 1 or 2 percent.

This options-related overstatement of earnings has exacerbated the already large valuation gap between technology and traditional blue chip companies.    As of last Friday's closing prices, the DJ Industrial Average was trading at 20.2x LTM earnings, versus 53.4x trailing earnings for the Nasdaq 100 index.    If you adjusted these figures to correct for the earnings overstatement due to not expensing employee option-related compensation, the P/E for the 30 companies in the DJIA would increase to 21.6x (assuming, conservatively, that expensing employee options would reduce LTM earnings by 7%).  However, for the 100 firms in the Nasdaq 100 index, the multiple of trailing earnings would increase to a  staggering 71.6x, assuming that excluding option expenses increased earnings on average by 34%.

(P/E's for the DJIA were taken from Barron's Market Laboratory (subscription required).   P/E's for the Nasdaq 100 are harder to come by since it is a market-capitalization weighted index, meaning that the relative weightings of its component companies change daily as stock prices fluctuate.  I calculated the P/E shown based on the February 26, 2004 closing index composition weightings and EPS data from MSN Money using an Excel external data function.)

The tech-heavy Nasdaq composite index has risen a torrid 82% to 2,029 from its post-bubble October 2002 low of 1,114.    While this is down 6% from its recent high of 2,153 on January 26th, it is still very high relative to the earnings and growth prospects of its constituent companies.    If the tech slump continues (and the Nasdaq composite has closed down for six consecutive weeks), look for hungry trial lawyers and ambitious state Attorneys General to revisit this issue of employee options accounting.