February 5, 2003

 

Ms. Suzanne Bielstein
Director of Major Projects and Technical Activities
Financial Accounting Standards Board
401 Merritt 7
P. O. Box 5116
Norwalk, CT 06856-5116

 

Re: File Reference No. 1102-001 - Accounting for Stock-Based Compensation: A Comparison of FASB Statement No. 123, Accounting for Stock-Based Compensation, and Its Related Interpretations, and IASB Proposed IFRS, Share-based Payment

 

Dear Ms. Bielstein:

 

The Financial Accounting Policy Committee (FAPC) of the Association for Investment Management and Research (AIMR)¹ is pleased to comment on the Financial Accounting Standards Board's (FASB) Invitation to Comment, Accounting for Stock-Based Compensation: A Comparison of FASB Statement No. 123, Accounting for Stock-Based Compensation, and Its Related Interpretations, and IASB Proposed IFRS, Share-based Payment [the "Comparison"]. The FAPC is a standing committee of AIMR charged both with maintaining liaison with standard setters who develop financial accounting standards and regulate financial statement disclosures and with responding to new regulatory initiatives. The FAPC also maintains contact with professional, academic, and other organizations interested in financial reporting.

 

General Comments

 

The FAPC supports the general tenor of this proposal to

 

…Solicit comments on certain issues that the Board will discuss when, in accordance with its objectives of improving U.S. financial accounting and reporting standards and promoting international convergence of high-quality accounting standards, it considers whether it should propose any changes to the U.S. accounting standards on stock-based compensation.

 

We believe that both objectives-improving U.S. reporting standards and promoting international convergence-are critically important steps toward improving the efficiency and effectiveness of global financial markets. Specifically, these will help to serve the needs of investors and other users for relevant, timely, reliable, transparent, and comparable information. These needs should supersede other considerations and we believe them to be the intent of financial reporting rule-makers.

 

The FAPC is pleased to observe that the similarities between Statement 123 and the IFRS relate to the most fundamental principles that underlie the two approaches. Consequently, we believe that, with careful consideration by the FASB and the IASB, the differences can be resolved satisfactorily to the benefit of all market participants.

 

Specific Comments

 

Issue 1: Statement 123 provides a scope exclusion for ESOPs and certain ESPPs, and the Proposed IFRS does not. Which view do you support and why? (Refer to page 19.)
The FAPC believes that stock-based compensation should be accounted for by a single method, regardless of the specific legal or contractual structure in which the awards are granted. That is, there should be a single method of accounting for all stock-based compensation. For example, we do not believe that there is a fundamental economic difference between stock awarded to employees through ESPPs and through executive compensation stock option plans. Rather, we believe that they differ more in the relative amounts of compensation awarded. Thus, in the economic sense, the various plans may be considered to represent different points on a single spectrum of awards because all, regardless of form of final payment,

 

• are based upon, or indexed to, the company's stock;
• offer favorable terms to the employees relative to outside shareholders; and
• involve an economic sacrifice to the company, either directly or indirectly, through the opportunity cost of foregone and/or expended capital.

 

We believe also that the principle of a single method of accounting for transactions that in essence are economically similar, is consistent with the intent expressed by the FASB Proposal, Principles-Based Accounting. Not only is there no discernable benefit to investors from a proliferation of accounting choices for similar transactions, but the practice entails considerable costs, both to

 

• preparers who must fully explain in the notes to the financial statements the methods applied and their relative effects; and
• users of the statements who must try to decipher the disclosures and extract the economic significance of the transactions.

 

Issue 2: In measuring the fair value of stock options granted to employees, both Statement 123 and the Proposed IFRS require use of an option-pricing model that takes into account six specific assumptions. The standards provide supplemental guidance for use in selecting those assumptions. (Refer to page 20.)

 

Issue 2(a): Do you believe that an accounting standard should mandate the use of an option-pricing model for measurement purposes? If not, what other approaches do you believe would provide more consistent and reliable estimates of the fair value of employee stock options granted and why? (Refer to page 21.)

 

Issue 2(b): If you agree that an accounting standard should mandate the use of an option-pricing model, do you believe that a particular model should be mandated? If so, which model should be required to be used and why? (Refer to page 21.)

 

Issue 2(c): If you agree that an accounting standard should not mandate the use of a particular option-pricing model, do you believe that additional disclosures should be made to improve the user's ability to compare the reported financial results of different enterprises? If so, what types of additional information should be required to be disclosed? (Refer to page 21.)
The objective of using an option-pricing model is to measure the fair value of the options awarded at grant date and, depending upon the measurement method eventually applied in a revised Statement 123 or the IFRS, possibly thereafter. The FASB believes that fair value is the only relevant measure for financial reporting purposes. We agree strongly with this position and believe that

 

• The best measure of fair value is determined by market-exchange prices.
  
  
• Failing the availability of such prices, fair values should be estimated using a model that is in widely accepted use in the markets for valuing such assets.
• The necessary inputs to the model also should be market based. That is, the inputs should be those used by market participants for valuing similar assets.

 

One of the most widely-used and accepted models in current market use for valuing stock options is the Black-Scholes model. Indeed, the model has been so widely adopted for valuing option-based contracts, and its theoretical grounding is such, that its developers were awarded a Nobel Prize.

 

Mandating that a single, specified model be used for the valuation of all stock options, along with clearly defined inputs to the model, ensures that the important attributes of consistency and comparability in disclosure are maintained across companies. On the other hand, relevance is by far the most important disclosure attribute. To the extent that a particular measurement method produces disclosures that are both more timely and accurate than an existing method, relevance will be enhanced. The FAPC recognizes that it is possible that other models may be developed in the future that supersede the Black-Scholes model (other models are currently available). It is highly probable that over time improvements will be made to the accepted models that better reflect the economic terms and conditions of particular contracts. Such research is ongoing at a rapid rate.

 

Consequently, the FAPC believes that it is essential to specify the characteristics the models should exhibit, for example, widely accepted use in the financial markets for valuing similar assets, with inputs constrained to those used in such models in the financial markets. Thus, we agree with the FASB and the IASB that the standards should not specify use of a particular model.

 

Issue 2(d): Statement 123 and the Proposed IFRS require that certain modifications be made to the outcome of an option-pricing model to address certain features of employee stock options. If you believe that other modifications should be made to improve the consistency and reliability of those outcomes, please describe those modifications and why they should be required. (Refer to page 21.)
The FAPC believes that the modifications, including adjustment for expected forfeiture rates and nontransferability, are sufficient at the present time for capturing the effects of specific conditions not explicitly addressed by the standard option pricing models. It is likely that measurement technology for these instruments will continue to improve and that the enhanced methods will include more efficient ways of adjusting for these conditions. We would hope that the final unified standard will be sufficiently robust to admit improvements in measurement technology.

 

Issue 2(e): Do you believe that additional guidance for selecting the factors used in option-pricing models is necessary to provide added consistency and comparability of reported results? If so, what types of guidance should be provided and in which areas? (Refer to page 21.)
As we state above, we believe that the factors and inputs should be those used and accepted widely for developing fair values in the financial markets. Such guidance is likely to provide the greatest consistency and comparability, as well as the most relevant and reliable valuations. However, we do not believe that company-specific or management-determined factors should be permitted to be used for the basic valuations.

 

It is apparent that expected forfeiture rates and the average expected term to exercise may be influenced to some extent by specific contractual terms and company conditions. Consequently, we believe that the company should be required to

 

• disclose both the expected and the actual or realized rates in the notes to the financial statements; and
• adjust the expected rates should they depart materially from the actual rates.

 

Issue 3: Do you believe that employee and nonemployee transactions are distinct and, therefore, warrant different measurement dates for determining the fair value of equity instruments granted? If so, why? If not, why not? (Refer to page 24.)
See our response to Issue 1 above. We believe that all stock-based awards granted as consideration for the receipt of goods and services, whether to employees or nonemployees, should be accounted for by the same rule. Differences in conditions or contingencies should be recognized in the measurement of the fair values at grant date, that is, under Concepts Statement 7, in the probability distributions assumed.

 

Issue 4: Do you believe that the fair value of equity awards granted to nonemployees that include performance conditions can be measured with sufficient reliability to justify a grant-date measurement method? If so, why? If not, why not? (Refer to page 24.)
See our responses to Issues 1 and 3 above. The FAPC fails to find a compelling reason to justify a different accounting treatment for nonemployee awards as compared to those for employees. Both types involve contingencies and other uncertainties as do most accounting measurements.

 

Issue 5: Do you believe the notion of issuance is conceptually of importance in the design of a standard on stock-based compensation? If so, why? If not, why not? (Refer to page 25.)

 

Issue 6: Do you believe an equity instrument subject to vesting or other performance conditions is issued, as defined by Statement 123, at the grant date? If so, why? If not, why not? (Refer to page 25.)
The FAPC has long held that executory contracts should be accounted for in the financial statements at inception of the contract, recognizing their anticipated effects on assets, liabilities, and the equity of the company. Our belief is based in the principle that the financial statements should fully reflect the fair values of all exchanges and transactions, including commitments and other arrangements that have, or possess the potential to have, an economic effect on the risks and rewards of the company.

 

In this context, issuance is the contractual milestone at which all material contingencies and terms have been fulfilled and completed by all parties to the stock-based contract or arrangement. That is, it is the point at which all uncertainties and risks are removed. This milestone is important, for example, in determining the final value of a stock-based award in which payment will be made in cash rather than the issuance of shares. We believe that executory contracts containing milestones should be accounted for, taking into consideration the economic effects for each milestone.

 

It is the view of the FAPC that at grant date, stock option awards are more properly termed "options on options," which are executory contracts, and should be accounted for as such at fair value beginning on the grant date, including the fair value of the conditional obligation. As the milestones-performance, vesting, or other conditions-are achieved over the period defined by the contract, the probability of issuance of stock (or the payment of cash in the case of stock-appreciation rights) will increase.

 

Consistent with our view, on vesting date, the option on an option is converted to a straight call option. At that point, the company has an unconditional obligation to award stock when the holder of the option "calls" the stock by exercising the option. Consequently, the FAPC believes that vesting and issuance are of importance in defining the shape of the probability distribution of the awards and must be taken into account in determining the fair value of the awards on grant date and going forward to exercise.

 

Issue 7: Do you believe that the effect of forfeiture should be incorporated into the estimate of fair value per equity instrument (IASB approach)? If so, why? If not, why not? (Refer to page 28.)

 

Issue 8: Should failure of an award holder to satisfy the conditions that entitle the holder to retain or receive the promised benefits affect the amount of compensation expense that should be recognized related to that award? If so, why? If not, why not? (Refer to page 28.)

 

Issue 10: Which of the two attribution methods described by the standards do you believe is more representationally faithful of the economics of stock-based compensation arrangements and why? (Refer to page 35.)

 

Issue 11: Statement 123 does not ascribe value to services received in exchange for equity instruments that are later forfeited (that is, recognized compensation expense is reversed upon forfeiture), whereas the Proposed IFRS ascribes value to such services through its units-of-service attribution method (that is, recognized compensation expense is not reversed upon forfeiture). If you support the Proposed IFRS's view, do you believe the units-of-service method ascribes an appropriate value to services received prior to forfeiture? If so, why? If not, why not? (Refer to page 35.)

 

Issue 12: Do you believe that the actual outcome of performance awards should affect the total compensation expense incurred by an enterprise? If so, why? If not, why not? (Refer to page 38.)

 

Issue 13: Do you believe that this issue is important in considering an attribution model's validity? If so, why? If not, why not? (Refer to page 40.)
We believe that the questions posed in Issues 7, 8, 10, 11, 12, and 13 are closely inter-related and therefore, will address them jointly.

 

The probability of forfeiture is one of the major factors that must be considered in determining the fair value of the award at grant date. If it is known at grant date that not all option awards will vest and be exercised, then a failure to take the probability of forfeiture into account in determining the fair value of the grant will result in an overstatement of the fair value of the award and thus, the compensation to be expensed. Retroactive restatement when the forfeitures occur is not a satisfactory solution and is not consistent with the principles of fair value accounting and measurement.

 

The question raised in Issue 9 goes to the heart of what the FAPC believes to be the major point of difference between the FASB's Statement 123 and the IASB's proposed IFRS. However, we do not believe that this issue is, or should be allowed to be, a major stumbling block in achieving convergence to a single high-quality standard. Because the two approaches can, under some circumstances, lead to materially different amounts of compensation recognized over the term of the awards; and different amounts within each period, the issue must be carefully examined and a satisfactory common solution developed. The FAPC believes that convergence can be achieved.

 

The FASB's and IASB's two different approaches result from two different theories regarding the award of stock options. As we understand it, the theory adopted by Statement 123 holds that at grant date the company and grantee enter into an executory contract with varying conditions and contingencies:

 

Until all such conditions and contingencies have been resolved and fulfilled, no binding commitment for the company to perform has been made, and it is not obligated to award, nor should it recognize, any compensation for goods or services received in the interim.

 

Since it is not known with certainty at grant date whether forfeiture will or will not occur, the Statement requires that a presumption be made at that time about the probability of forfeiture. The total fair value of compensation at grant date is then adjusted for the probability of forfeiture and allocated over the conditional or performance periods until such contingencies no longer exist. Should failure to perform occur, resulting in forfeiture prior to issuance, Statement 123 requires that such compensation recognized in the financial statements be adjusted, or "trued-up," to the actual nonforfeiture rates in order to recognize compensation based upon the revised amounts of stock expected to be issued. Thus, the theory applied in Statement 123, "modified grant date accounting," focuses accounting and the recognition of compensation expense over the period based upon the ultimate issuance of stock.

 

The IASB's theory leads to a different accounting result and different compensation recognized. The IFRS states that the purpose of share-based payment recognition is to reflect the

 

…Fair value of goods and services expected to be received over the contractual period, measured at grant date, taking into account the estimated probability of forfeiture.

 

Thus, the measurement focus is shifted from the actual issuance of stock, and the fair value measurement of that issuance at grant date and following until uncertainties are removed under Statement 123, to the measurement of the fair value of the goods and services received prior to actual forfeiture under the IFRS.

 

Under Statement 123, if the employee leaves the company, forfeiting the right to receive stock, that event is recognized by adjusting estimated compensation previously recorded to the amount that reflects the actual forfeitures. Under the proposed IFRS, any compensation recognized prior to forfeiture is considered to have been the fair value of goods and services the company received in a fair exchange and is not adjusted. The logical conclusion of this theory is that as a result of the voluntary or involuntary forfeiture of the stock award, the employee has effectively contributed or donated the fair value of the compensation to the company. Consequently, no adjustment should be made to reflect the actual amount of stock eventually to be awarded, measured at fair value at grant date.

 

Note that accounting for compensation (or the value of goods and services) ceases for both Statement 123 and the IFRS at forfeiture. Thus, the real difference between the two methods is whether or not it is appropriate to true-up, reflecting the actual as compared to estimated forfeiture.

 

The FAPC believes that the direct implication of vesting conditions, including performance requirements, is that the employee is expected to provide future services to the company over the vesting period. Under this assumption, the conditional award is granted in the anticipation that the company will receive benefits from the continued performance by the employee.

 

It is possible that this reasoning may provide ground for the development of a hybrid approach acceptable to all parties. The FAPC believes that, to the extent vesting or performance conditions are associated with stock option awards, a strong (rebuttable) presumption must exist that the economic essence of those conditions is to require and reward future service and performance of value to the company and its owners. That is, future service is being exchanged by the employee for future awards of stock. Consequently, the natural conclusion is that the compensation expense, as measured at fair value at grant date, should be allocated ratably over the vesting period until such time as the employee voluntarily or involuntarily forfeits the right to the compensation, consistent with the provisions of the proposed IFRS.

 

Issue 9: Do you agree that the result of the IASB's approach to calculate the fair value of equity instruments of nonpublic entities would be closer to fair value than minimum value? If so, why? If not, why not? (Refer to page 29.)
The FAPC agrees. We believe that fair value is the best measure of all such grants and that minimum value is not consistent with the fair value objective.

 

Issue 14: Do you believe that the measurement-date criteria in Issue 96-18 accurately reflect the economics of transactions with nonemployees? If not, why not? (Refer to page 43.)
As stated above, we believe that a single, consistent method of accounting for stock-based compensation should be followed for all grants of such compensation, regardless of the recipient of the awards, or the legal or contractual structure of the award. Consequently, we do not agree that different measurement-date criteria should be used for transactions with nonemployees. However, the fair value measurement at grant date should factor in the probability of nonperformance. In addition, contractual milestones would be of particular importance in deciding when a contractual commitment has been entered into, and thus, when a contingent obligation exists. For example, many such contracts make provision for payments or other damage awards upon failure to perform to a particular interim stage or completion.

 

Issue 15: Do you believe that all of the tax benefits derived from stock-based compensation arrangements should be recognized in the income statement? If so, why? If not, why not? (Refer to page 46.)
We believe that all tax benefits resulting from stock-based compensation awards should be recognized in the income statement. We believe that this treatment is consistent with the provisions of Statement 109.

 

Issue 16: As discussed in paragraph 83 of this Invitation to Comment, the Proposed IFRS expands on the disclosure requirements in Statement 123. Do you believe that those expanded disclosures would be more informative to users of financial statements? If so, why? If not, why not? (Which of the disclosure requirements should be eliminated or modified in that case?) (Refer to pages 47 and 48.)

 

Issue 17: Please describe any additional disclosures that you believe should be required in order to inform a user of financial statements about the economics of stock-based compensation arrangements. (Refer to page 48.)
The FAPC believes that the disclosures required in Statement 123, as well as the additional disclosures that would be required by the proposed IFRS, are essential. The disclosures should include any contractual terms, conditions, and other arrangements associated with the awards, including clear discussion of any performance expectations or conditions.

 

Consistent with our views on fair value reporting, disclosures should provide sufficient information to enable users to understand

 

• the valuation models used;
• the inputs to the models;
• the sources of data used for the inputs; and
• any estimates developed, either as inputs for the models or as adjustments to the fair values obtained from the models.

 

Additional disclosures that the FAPC believes are crucial to an understanding of the potential effects of the options on the company's operations include:

 

• the current market value of all outstanding options;
• a discussion of the vesting conditions;
• a discussion of the probability of those conditions being met;
• a sensitivity analysis of the effects of critical valuation assumptions;
• a disclosure of the basis for adjusting historically observed input, for example, volatility.

 

For many companies, stock-based compensation is a highly material expense, possibly the largest single expense. Consequently, we believe sensitivity analysis should be disclosed for the critical assumptions or inputs made, for example, the measure of volatility used in the valuation model. Indeed, the FAPC believes that sensitivity analysis should be routinely disclosed for any critical assumptions or value drivers in the company.

 

Concluding Remarks

 

The Financial Accounting Policy Committee appreciates the opportunity to express its views on the Board's Proposal: Accounting for Stock-Based Compensation: A Comparison of FASB Statement No. 123, Accounting for Stock-Based Compensation, and Its Related Interpretations, and IASB Proposed IFRS, Share-based Payment. If the Board or staff have questions or seek amplification of our views, please contact Rebecca McEnally at 1-434-951-5319 or at rebecca.mcenally@cfainstitute.org. We would be pleased to answer any questions or provide additional information you might request.

 

Respectfully yours,

 

Jane Adams Chair, Financial Accounting Policy Committee

Rebecca McEnally, Ph.D., CFA Vice-President, Advocacy, AIMR

 

¹ With headquarters in Charlottesville, VA, and regional offices in Hong Kong and London, the Association for Investment Management and Research is a non-profit professional organization of 61,600 financial analysts, portfolio managers, and other investment professionals in 113 countries of which 49,200 are holders of the Chartered Financial Analyst (CFA ) designation. AIMR's membership also includes 119 affiliated societies and chapters in 29 countries. AIMR is internationally renowned for its rigorous CFA curriculum and examination program, which had more than 100,000 candidates from 143 nations enrolled for the June 2002 exam.