GSA Corporate Governance

Stock Option Expensing: A Threat to the Fabless Industry

GSA Board of Directors
Jodi Shelton, GSA Executive Director
October 2002

The GSA represents emerging businesses as well as small- and medium-sized public and private companies. The primary group represented by GSA is the fabless semiconductor segment. "Fabless," simply stated, defines a semiconductor company that does not own or operate its own manufacturing facility, also known as a "fab." The fabless semiconductor outsourcing business model was established in the 1980s and proliferated in the 1990s. Currently more than 500 fabless semiconductor companies operate in North America. Furthermore, even larger, established semiconductor companies have adopted this business model because of the onerous costs of building a fab and the need to focus on designing more innovative semiconductor solutions.

The fabless segment is comprised of start-up companies that are in the idea formation or product definition mode, pre-IPO companies and larger, established public companies. Some of the most successful fabless companies such as Xilinx, Broadcom, Nvidia, Qualcomm CDMA Technologies, ATI Technologies and Altera have annual revenues near or in excess of $1 billion. Because of the relatively small size of fabless companies, they are heavily dependent upon employee stock options (ESOs) to attract and retain talent, and to date it has worked! The fabless industry is a key example of value creation for the U.S. economy over the last decade. Fabless companies have pioneered several of the most lucrative end markets for semiconductors, including the FPGA market, networking processors, graphics and communications. Fabless companies are responsible for many of the industry's most innovative products. They have also been the fastest-growing group within the semiconductor segment. Fabless growth has outpaced the growth of the semiconductor industry in both up and down markets. Also, three fabless companies have held the honor of the fastest-growth semiconductor companies to achieve $1 billion in annual sales - Cirrus Logic, Broadcom and Nvidia.

GSA believes that fabless companies will be the group most negatively affected by mandatory expensing of ESOs because employees of fabless companies are traditionally paid less in salary and expected to work more hours, with the motivating factor being the distribution of stock options. The idea is that hard work, combined with innovative products and a growth market with the right vision may result in personal wealth for these employees. If ESO expensing were mandatory, there would be less opportunity for companies to offer options, meaning hard-working employees would reap less reward for their dedication and, therefore, be less motivated to innovate. U.S. innovation could be stifled without the continuing development of breakthrough products that this segment continues to drive.

Most fabless companies are private, with the ambition of becoming public. Any action that might affect a company's profitability jeopardizes the potential for their initial public offerings (IPOs). Private fabless companies face a competitive environment, where they are fighting to retain experienced design staff. Therefore, to build their companies, entrepreneurs must be allowed to offer stock options to prospective employees as an enticement for individuals to invest themselves in a company. And the entrepreneurs who create companies should be rewarded for the risk of leaving established industry positions to create opportunities for others. The typical salary of a private fabless CEO ranges between $150,000 and $200,000. Fabless CEOs are taking less compensation to preserve cash in hopes that they will create viable, successful companies. In the end, their ultimate reward, and that of their employees, will be based on the success of the company - the formula every investor wants!

It is apparent that ESO expensing would be a direct and painful assault on fabless employees, that, until now, have shared in the ownership of companies created by their hard work. Ironically enough, most pundits agree that although there would be a major retreat in the distribution practice of ESOs to non-executives if expensing were to occur, executives would still be granted options, or would be otherwise compensated.

Emerging economies understand the value of equity incentives and have adopted this tool to build technological prowess. Specifically, Taiwan and Mainland China have been aggressive in their distribution of stock options to employees as compensation.

The GSA bases its strong opposition to legislation and accounting changes that would treat broad-based ESOs as an expense on the fact that there is no accurate methodology for valuing ESOs, and expensing them would distort profitability and mislead investors.

It is impossible to place an accurate value on an option in the year it is granted, since it may be exercised for some unknown price at an indeterminate date, or may not be exercised at all. Stock options give one the right, but not the obligation to buy stock at some point in the future for a set price. There is no reliable way to predict when, or if, an employee will exercise the option and what value it will have at that time. The options typically vest over four years, and the employee must continue employment with the company in order to vest. We all know now, more than ever, how volatile the stock market can be. This volatility has placed many fabless employees in a situation where options that were granted some time during the last five years are worth less today than when the option was originally granted.

For example, suppose an engineer was granted an option to buy 1,000 shares of a company's stock for $80 per share in 2000 (as of that date in 2000, the share price was $40) on a four-year vesting schedule. In 2002, he is two years vested; however, the company's shares are at $10 per share. How should the expensing of these ESOs be handled? In one scenario, the expensing charge would be amortized over the vesting period, and the company would be charged $10,000 per year against earnings over the vesting period. The company would record that $10,000 per year as compensation. The company's profit would decrease by this amount, but no cash would change hands. This employee cannot get any value from these options. So what happens when reality differs from this formula? Are they restated, and if so, how often and in what way?

Including an unreliable estimate of the fair value of options in a company's income statement would distort earnings. The potential overstatement of the options' economic cost in the financial statement would definitely curtail their use. It is unwise to put a presumably faulty estimate of a future cost into a current income statement or to reverse it when the fault is realized.

Despite what seems like a progressive grass roots program, public opinion appears to favor ESO expensing and, thus, may influence congressional action and/or accounting rule changes. But this opinion is misdirected. Investors are dissatisfied with overall corporate governance abuses by such companies as Enron, Adelphia and Worldcom. The issue of ESO expensing has been improperly confused with overt abuses in executive compensation.

The GSA urges its member companies and their employees to be proactive and vocal in opposing ESO expensing. GSA will be working hard with other organizations to influence FASB. But we must also work to change public opinion by identifying the real issues and getting the truth out. Please help - write letters, talk to friends and vote.