Re:
|
Analog Devices, Inc. | |
Form 10-K for the year ended October 29, 2005 | ||
Filed November 21, 2005 | ||
SEC File No. 001-07819 | ||
1. | On page 16 you state that the 9% decline in revenue between 2005 and 2004 is due to an inventory correction at your customers. You also state that the decline was due to a decrease in orders and weaknesses in certain markets. On page 17 you state that the 29% increase in revenue between 2004 and 2003 was due to an increase in demand. We also note from your Form 8-K dated November 15, 2005 that you experience normal seasonality in the consumer and industrial markets. Please revise MD&A in future |
filings to include (a) an analysis explaining the underlying reasons for significant change noted, (b) a discussion of the extent to which such increases are attributable to increases in prices or to increases in the volume or amount of goods and services being sold or to the introduction of new products or services, (c) seasonal trends or variations in sales, (d) shipments of product at the end of a reporting period that significantly reduced customer backlog and that reasonably might be expected to result in lower shipments and revenue in the next period, (e) changing trends in shipments into, and sales from, a sales channel or separate class of customer that could be expected to have a significant effect on future sales or sales returns, and (f) a quantification for each significant factor noted when you identify the contribution of two or more factors as the cause for material changes. Please also revise MD&A in future filings to describe any known trends or uncertainties that have had or that you reasonably expect will have a material favorable or unfavorable impact on net sales or revenues or income from continuing operations. Please see Item 303 of Regulation S-K, SAB Topic 13.B, FRC 501, and Release 33-8350. | ||
Response: | ||
Because the MD&A included in the Form 10-K and Form 10-Q are comparative year-to-year or period to period disclosures the issue of seasonality does not ordinarily arise. However, where we explain sequential quarter-to-quarter revenue variations and seasonality is deemed to be a factor we do comment appropriately. For example, in your comment, you reference our Form 8-K dated November 15, 2005 where we indicated we have normal seasonality in the consumer and industrial markets. In that filing we were explaining why we believed the revenue in our first fiscal quarter of 2006 would be approximately equal to the immediately prior quarter. Typically, our first quarter is weak as a result of the impact of holiday shutdowns particularly in Europe and the US. | ||
We use the US dollar value of revenue as the principal financial metric when managing our business because the selling prices and unit volumes fluctuate as a result of the mix of products sold. During any quarter we ship over 300 million individual units to our customers and distributors throughout the world and sell thousands of different products to tens of thousands of customers. Therefore, our MD&A includes a discussion of changes in revenue for product categories, geographic sales, end markets, etc. To the extent appropriate, we will expand future MD&A disclosures to include the items you addressed above. |
2. | We see from the disclosure on page 47 that management changed the method for calculating volatility in determining the value of employee stock options to implied volatility, which decreased volatility from 69.2% in fiscal year 2004 to 27.4% in fiscal year 2005. Please revise to include an explanation of the method used to estimate the expected volatility of your share price. Your explanation should include (a) why you changed your method of determining estimated volatility and (b) why you now only use implied volatility and not a combination of historical volatility and implied volatility. |
You should consider summarizing your evaluation of the factors listed in Questions 2 and 3 of SAB Topic 14.D.1 as part of this disclosure. Please refer to Questions 5 of SAB Topic 14.D.1, SEC Release No. FR-60 and Section V, Critical Accounting Estimates, in SEC Release No. FR-72. | ||
Response: | ||
Prior to fiscal 2005, we used historical volatility for the period commensurate with the expected term of our options as the estimate of expected volatility used in the Black-Scholes valuation model to calculate fair value. Given the unprecedented increase and subsequent decline in stock prices in 1999 and the years following (see below), we do not believe that historical volatility is the most reasonable measure of expected volatility. |
FAS 123R and SAB 107 provide guidance that states implied volatility may be a better estimate of expected volatility in cases where the company does not believe it is reasonable to assume that past stock price behavior is likely to be indicative of expected future stock price behavior. Both FAS 123R and the SAB 107 permit the exclusive reliance on implied volatility if certain conditions are met. We believe we have met the criteria as addressed in question 3 and 4 of Topic 14 and have included additional discussion with our response to your comment number four below. | ||
We have added stock-based compensation to the Critical Accounting Policies and Estimates section of our quarterly filing on Form 10-Q. The revised disclosure includes the method used to estimate expected volatility, why we changed our method of estimating expected |
volatility and why we are now exclusively relying on implied volatility to estimate expected volatility. | ||
We have incorporated our response to your comment in the following disclosure included in our Form 10-Q for the quarter ended January 28, 2006 filed on February 15, 2006: | ||
Expected volatility we are responsible for estimating volatility and have considered a number of factors, including third-party estimates, when estimating volatility. For options granted prior to fiscal 2005, we used historical volatility to estimate the grant-date fair value of stock options. We changed our method of estimating expected volatility for all stock options granted after fiscal 2004 from exclusively relying on historical volatility to exclusively relying on implied volatility. This change was the result of a thorough review we undertook which included consultations with several third party advisors. We currently believe that the exclusive use of implied volatility results in a more accurate estimate of the grant-date fair value of employee stock options because it more appropriately reflects the markets expectations of future volatility. Historical volatility during the period commensurate with the expected term of our stock options over the past several years included a period of time that our stock price experienced unprecedented increases and subsequent declines. We believe that this past stock price volatility is unlikely to be indicative of future stock price behavior. Options in our stock are actively traded on several exchanges. Implied volatility is calculated for the period that is commensurate with the options expected term assumption. Because this term often exceeds the period for which there are exchange traded options in our stock, statistical techniques are used to derive the implied volatility for traded options with terms commensurate with the options expected term of five years. This calculation of implied volatility is based on the most recent five-day period of trades of our exchange-traded options as of the date of grant. In general, the higher the expected volatility used in the Black-Scholes valuation model, the higher the grant-date fair value of the option. |
3. | We note your reference to a third-party estimate. Please note that if you elect to continue to make such a reference and intend to incorporate your Form 10-K by reference in any registration statement you will be required to identify the appraisal firm under Experts and include their consent in the registration statement. Alternatively, we encourage you to instead clearly disclose that management is primarily responsible for estimating volatility. We will not object if you wish to state, in future filings, that management considered a number of factors, including third-party estimates, when estimating volatility. Regardless of your decision, your disclosure should clearly indicate that management is responsible for the estimate. Please revise future filings as appropriate. |
Response: | ||
We have revised our disclosure in our quarterly filing on Form 10-Q to more clearly reflect that we are responsible for the estimate of expected volatility. | ||
We have incorporated our response to your comment in the following disclosure included in our Form 10-Q for the quarter ended January 28, 2006 filed on February 15, 2006: | ||
Expected volatility the Company is responsible for estimating volatility and has considered a number of factors, including third-party estimates, when estimating volatility. For options granted prior to fiscal 2005, the Company used historical volatility to estimate the grant-date fair value of stock options. The Company changed its method of estimating expected volatility for all stock options granted after fiscal 2004 from exclusively relying on historical volatility to exclusively relying on implied volatility. This change was the result of a thorough review the Company undertook which included consultations with several third party advisors. The Company currently believes that the exclusive use of implied volatility results in a more accurate estimate of the grant-date fair value of employee stock options because it more appropriately reflects the markets expectations of future volatility. Historical volatility during the period commensurate with the expected term of the Companys stock options over the past several years included a period of time that the Companys stock price experienced unprecedented increases and subsequent declines. The Company believes that this past stock price volatility is unlikely to be indicative of future stock price behavior. Options in the Companys stock are actively traded on several exchanges. Implied volatility is calculated for the period that is commensurate with the options expected term assumption. Because this term often exceeds the period for which there are exchange traded options in the Companys stock, statistical techniques are used to derive the implied volatility for traded options with terms commensurate with the options expected term of five years. This calculation of implied volatility is based on the most recent five-day period of trades of the Companys exchange-traded options as of the date of grant. |
4. | We note that you have placed exclusive reliance on implied volatility in providing your estimate of expected volatility. So that we may better understand how your selection meets the stated objectives of SFAS 123R and SAB Topic 14, please tell us how you considered the following in your evaluation: 1) the volume of market activity of the underlying shares and traded options; 2) the ability to synchronize the variables used to derive implied volatility; 3) the similarity of the exercise prices of the traded options to the exercise price of the employee share options; and 4) the similarity of the length of the term of the traded and employee share options. See Question 3 of SAB Topic 14. Please also address the extent to which the following factors are present: |
| You use a valuation model that is based upon a constant volatility assumption to value your employee share options. Please note that a company that utilizes an option pricing model that incorporates a range of expected volatilities over the options contractual term should consider the factors listed in SFAS 123R and those discussed in Questions 2 |
and 3 in SAB Topic 14 to determine the extent of your reliance (including exclusive reliance) on the derived implied volatility. | |||
| Your implied volatility is derived from options that are actively traded. Please explain. | ||
| The market prices (trades or quotes) of both the traded options and underlying shares are measured at a similar point in time to each other and on a date reasonably close to the grant date of the employee share options. | ||
| The traded options have exercise prices that are both (a) near-the-money and (b) close to the exercise price of the employee share options. | ||
| The remaining maturities of the traded options on which the estimate is based are at least one year. | ||
See Question 4 of SAB Topic 14. | |||
Response: | |||
SFAS 123R and SAB Topic 14 provide guidance which states that implied volatility may be a better estimate of expected volatility in cases where the company does not believe it is reasonable to assume that past stock price behavior is likely to be indicative of expected future stock price behavior. Both SFAS 123R and Topic 14 permit the exclusive reliance on implied volatility if certain conditions are met. We believe we have met the criteria as addressed in question 3 and 4 of Topic 14 and have addressed each of these topics below in response to the four items listed in question 3 of Topic 14. We are responsible for calculating our implied volatility and have engaged a third party to assist with this calculation. The third party utilizes the following information in their calculation of our implied volatility: |
1. | The volume of market activity of the underlying shares and traded options |
i. | Our stock is actively traded on the NYSE. The average volume of these trades over the last three months was approximately 3.2 million shares per trading day. Options in our stock are also actively traded on several exchanges including: AMEX; BOX; CBOE; ISE; PCX; and PHLX. |
2. | The ability to synchronize the variables used to derive implied volatility |
i. | The traded options and underlying shares are measured at a similar point in time to each other and on a date reasonably close to the grant date of the employee share options. The volatility is derived from the closing prices of our stock and exchange traded options from the most recent five trading days prior to the grant date of the employee stock option. We use a valuation model (the Black Scholes Merton Formula) that is based upon a constant volatility assumption to value our stock options. |
3. | The similarity of the exercise prices of the traded options to the exercise price of the employee share options |
i. | All traded options are used to calculate our implied volatility. These traded options have exercise prices that are both near-the-money and out-of-the-money. The near-the-money options were given more weight in the calculation of our implied volatility because they are more actively traded and the exercise prices of these options are closer to the exercise prices of the employee stock options. |
4. | The similarity of the length of the term of the traded and employee share options |
i. | Implied volatility is calculated for the period that is commensurate with the options expected term assumption of five years. Because this term exceeds the period for which there are exchange traded options in the Companys stock, statistical techniques are used to estimate the stochastic volatility model that best fits the prices of the exchange traded options that do exist. This model is then used to derive the implied volatility for traded options with terms commensurate with the options expected term of five years. Footnote 49 of Topic 14 reads The staff believes it may also be appropriate to consider the entire term structure of volatility provided by traded options with a variety of remaining maturities. If a company considers the entire term structure in deriving implied volatility, the staff would expect a company to include some options in the term structure with a remaining maturity of six months or greater. Based on this guidance, we have included all exchange traded options in the calculation of our implied volatility and weighted the longer maturities more heavily than those with shorter maturities for purposes of calculating our implied volatility. For example, during the five day time period prior to our December 6, 2005 option grant approximately 28% of our exchange traded options had terms longer than one year. These options were given a weight of 97% and our options with maturities less than one year were given a weight of 3% in the calculation of our implied volatility. Additionally, the staff noted in footnote 50 of Topic 14 that ...implied volatility derived from a traded option with a term of one year or greater would typically not be significantly different from the implied volatility that would be derived from a traded option with a significantly longer term. |
5. | We see that you recognized $20 million in fiscal year 2005 for severance and fringe benefit costs related to the closing of the California wafer fabrication operation. You disclose that none of the employees had been terminated as of the end of fiscal year |
2005. Under paragraph 11 of SFAS 146, if the employees are required to provide future service to receive the benefit, the liability should be recognized over the future service period. Please clarify, in your response and in future filings, the responsibility of the terminated employees to render service until they are terminated to earn the severance benefits and discuss how this affected your accounting treatment. | ||
Response: | ||
We believe the relevant accounting literature to account for the severance and fringe benefit costs related to our 2005 action is SFAS 88 and SFAS 112 because we have ongoing benefit arrangements for those employees to be terminated in connection with the closing of the California wafer fabrication operation. The severance benefit for these employees is calculated pursuant to our ongoing benefit plan as published in our HR Policy Manual, which is available to all employees. All employees for which amounts were accrued were subject to our published ongoing benefit plan. Accordingly, each employee for which severance benefits have been accrued as of October 29, 2005 can determine their severance benefit at any point in time. Further, the severance benefits accrued are consistent with our past actions taken in 2001, 2002 and 2003. SFAS 146 addresses the financial accounting and reporting for costs associated with exit or disposal activities, and paragraph 2(a) indicates that those costs include, but are not limited to: |
Before recording any severance costs, we identify specific employees by department and geographic location. For those employees subject to our ongoing benefit plans and statutory minimum requirements severance benefits are required to be accrued at the commitment date in accordance with SFAS 88 and SFAS 112. We have applied the provisions under paragraph 15 of SFAS 88, which state that an employer that provides contractual termination benefits must recognize a liability and a loss when both of the following conditions are met: It is probable that employees will be entitled to the benefits, and the amount can be reasonably estimated. Accordingly, since the termination benefits are contractually owed under our policies for involuntary terminations and do not involve an exchange offer and we had sufficient detail to make a reasonable estimate of the severance benefits to be paid, we accrued the severance benefits on the commitment date utilizing the event approach. In those instances, if any, in which the severance benefit to be provided is in excess of our published benefit policies, we record that excess benefit in accordance with SFAS 146. | ||
We have incorporated our response to your comment in the following disclosure included in our Form 10-Q for the quarter ended January 28, 2006 filed on February 15, 2006: |
Closure of Wafer Fabrication Facility | ||
During the fourth quarter of fiscal 2005, the Company recorded a special charge of $20 million as a result of a decision to close its California wafer fabrication operations and transfer production to facilities located in Massachusetts and Ireland, as well as to third-party wafer fabricators. The charge was for severance and fringe benefit costs that were recorded pursuant to SFAS 88, Employers Accounting for Settlements and Curtailments of Defined Benefit Pension Plans and for Termination Benefits, under the Companys ongoing benefit plan for 339 manufacturing employees and 28 general and administrative employees. The severance benefit is calculated based on length of past service, and employees must continue to be employed until they are involuntarily terminated in order to receive the severance benefit. As of January 28, 2006, all affected employees remain employed by the Company. The employment of most of the employees is expected to terminate upon the closure of the wafer fabrication facility which is planned for October of 2006. In addition to the charge recorded in the fourth quarter of fiscal 2005, the Company recorded additional expense in the first quarter of fiscal 2006, which consisted of $5.4 million of non-cash cost of sales expenses for additional depreciation due to shortened useful lives of certain manufacturing equipment and $0.5 million for stay-on bonuses. In accordance with GAAP, the Company expects to incur additional expenses related to this action during fiscal 2006 of approximately $23.6 million, of which approximately $16.2 million will be for non-cash cost of sales expense for additional depreciation, approximately $1.4 million will be for stay-on bonuses and approximately $6 million will be for estimated lease termination and clean-up costs. The closure of these facilities is expected to be completed by the end of fiscal 2006. |
6. | We see your disclosure that you have provided for potential tax liabilities due in various jurisdictions. To the extent that the amount is material to your financial statements, please tell us and revise future filings to disclose the amount recorded as a liability for uncertain tax positions. Refer to SFAS 5. | |
Response: | ||
Our accrued tax liability of $172,276,861 at October 29, 2005 includes the following: |
$41,680,991 | Current US, state and foreign tax filings. | |||
$126,423,243 | Estimated liability resulting from the IRS review of our fiscal 2001-2003 income tax returns. Recently, during February 2006, the IRS issued proposed adjustments on Form 5701 for the 2001-2003 IRS exam. The total estimated liability which would result from the issues described in the Form 5701 would be approximately $123,239,823. This amount includes the estimated liabilities for the fiscal 2001-2003 years under examination and the liabilities associated with the |
amended returns for fiscal 2004-2005 which we would be obligated to file. It also includes the liability for penalties. | ||||
$4,172,627 | Estimated state tax liabilities resulting from tax audits. The majority of this liability is the result of the issues from the IRS audit which will require amended state tax returns. |
The IRS is in the final stages of their exam and expect to determine a final settlement in the near future. Once the final settlement is determined, we will account for differences, if any, between the actual tax liability and the amounts accrued. | ||
As requested, in future filings, we will disclose any amounts recorded as a liability for uncertain tax positions. |
| The company is responsible for the adequacy and accuracy of the disclosure in the filing; | ||
| Staff comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the filing; and | ||
| The company may not assert staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States. |
Sincerely, |
||||
/s/ Joseph E. McDonough | ||||
Joseph E. McDonough | ||||
Vice President-Finance and Chief Financial Officer | ||||