Zep Inc.
August 11, 2011
Filed via EDGAR
Mr. Rufus Decker
Accounting Branch Chief
United States Securities and Exchange Commission
Washington, DC 20549
Re: Form 10-K for the Year Ended August 31, 2010 filed November 8, 2010
Form 10-Q for the Fiscal Quarter Ended May 31, 2011 filed July 7, 2011
File No. 1-33633
Dear Mr. Decker:
Zep Inc., a Delaware corporation (the “Company”, “we”, “us”, or “our”), is transmitting for filing the Company’s responses to the comments of the staff of the Division of Corporation Finance (the “Staff”) of the Securities and Exchange Commission (the “Commission”) contained in your letter dated July 29, 2011.
For convenience of reference, each comment contained in your July 29, 2011 letter is reprinted below in italics, numbered to correspond with the paragraph numbers assigned in your letter, and is followed by our corresponding response. Our responses also address matters discussed during our August 4, 2011 telephone conversation with the Staff.
Form 10-K for the Year Ended August 31, 2010
General
1. Where a comment below requests additional disclosures or other revisions to be made, please show us in your supplemental response what the revisions will look like. These revisions should be included in your future filings, including your interim filings, if applicable.
RESPONSE: We have provided, where applicable, the additional disclosures or other revisions in our responses below and will include these revisions and additional disclosures in our future filings, where applicable.
Financial Statements
Notes to the Financial Statements
Note 2 - Summary of Significant Accounting Policies
Goodwill and Identifiable Intangible Assets, page 53
2. We note your response to prior comment 3. You disclose that the customer relationship intangible assets recorded in connection with your acquisition of the North American operations of Waterbury Companies, Inc. are being amortized over a period ranging from 8 to 22 years. Given that only a small percentage of these intangible assets are amortized over the lower part of this range, please further clarify in your disclosures the amount of assets amortized over each end of the range or provide a table with a breakdown of the total amount of assets recorded by amortization period.
RESPONSE: The Company will expand this disclosure in its future filings with the Commission as follows:
“The value allocated to Waterbury’s customer relationships will be amortized on a straight-line basis ranging from eight to 22 years. The weighted average useful life of Waterbury’s acquired customer relationships is 21 years, and we expect to recognize amortization expense of approximately $1.3 million during each of the next five years in connection with these assets.”
We understand the estimation of the fair value of acquired identifiable intangible assets and the related determination of useful lives involves subjective and complex judgments. We advise the Staff that we will include additional disclosure regarding these judgments in the “Critical Accounting Estimates” section of our MD&A.
3. You provide an attrition summary which shows the attrition rates and useful lives by product line. Please tell us how you arrived at the useful lives based on these attrition rates.
RESPONSE: We determined the fair value of the customer relationships related to the Amrep, Inc. (“Amrep”) and Waterbury Companies, Inc. (“Waterbury”) acquisitions using the excess earnings method under the income approach. The income approach is a common valuation method acknowledged by the Financial Accounting Standards Board (“FASB”) in ASC 820-10-20. The income approach also helped us determine the useful lives of acquired customer relationships as it considers the
periods of expected cash flows and other pertinent factors in ASC 350-30-35-3. A key input into this income-based approach includes the attrition rates derived from historical sales data generated by the acquired customer base. Accordingly, attrition rates influence both the useful lives and the fair value of our acquired customer relationships.
We collaborated with an international, highly credentialed transaction and valuation advisory firm to assist with performing these analyses due to their inherent complexity and subjectivity. The advisory firm helped us to understand both industry best practices and generally accepted valuation methods for acquisitions of this nature. We consulted with this advisory firm to receive an objective understanding of appropriate valuation assumptions, including the point in time constituting when substantially all economic benefit has been derived from a customer relationship.
In summary, the excess earnings method can be used to determine the overall fair value of an intangible asset by estimating the amount of discounted residual (or excess) cash flows generated by the asset. The excess earnings method yields multi-year financial projections or cash flow “run-outs”. For the purposes of valuing acquired customer relationships, these projections begin with a total sales amount based upon historical sales information. Each year, we applied a consistent growth rate to the preceding year’s sales total reflecting our expectation that revenue from acquired/repeat customers will grow. We have stated in our external communications that we believe we conduct business in an industry that grows at approximately the rate of US Gross Domestic Product. To that end, our revenue growth assumptions in both the Waterbury and Amrep excess earnings models were set at 1.5%. We recognize that higher growth factors would result in longer useful life periods, but we believe our estimates to be conservative and appropriate. We did not include projected synergies between us and our acquired companies when utilizing the excess earnings method.
The sales in our product-line specific excess earnings models dissipate each year in accordance with the product-line specific attrition rates calculated from Amrep and Waterbury’s historical sales data. These revenue attrition factors more than offset the above-mentioned revenue growth factors. Expenses are applied against the acquired sales base that, over time, diminishes, and the resulting after-tax “profit” is discounted to the present. After a number of years, an acquired product line’s net economic benefit becomes fully exhausted (trends to a zero balance).
We base the remaining economic useful lives of customer relationships at the point in time we expect to realize substantially all of the benefit, or approximately 95%, of our acquired product lines’ projected cash flows. Therefore, our useful life determinations
are actually shorter than the economic benefit realization periods set forth in our excess earnings models.
To clarify further for the Staff our use of these excess earnings models, we have provided the following illustration, whose assumptions and mechanics are consistent with those used in our actual valuations and useful life determinations:
Income Approach
|
|
|
| Year 1 |
| Year 2 |
| Year 3 |
| Year 18 |
| Year 30 |
| |||||||
Total Customer Revenue |
| $ | 10,000 |
| $ | 10,150 |
| $ | 10,302 |
| $ | 10,457 |
| $ | 13,073 |
| $ | 15,631 |
| |
GDP-Based Revenue Growth |
|
|
| 1.5 | % | 1.5 | % | 1.5 | % | 1.5 | % | 1.5 | % | |||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||
Revenue from Repeat Customers |
| $ | 10,000 |
| $ | 10,150 |
| $ | 10,302 |
| $ | 10,457 |
| $ | 13,073 |
| $ | 15,631 |
| |
Attrition Factor (4% per annum) |
| 100.0 | % | 96.0 | % | 92.2 | % | 88.5 | % | 48.0 | % | 29.4 | % | |||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||
Revenue From Current Relationships |
| $ | 10,000 |
| $ | 9,744 |
| $ | 9,495 |
| $ | 9,251 |
| $ | 6,270 |
| $ | 4,593 |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||
Expenses, Other Charges, Taxes |
| $ | (9,750 | ) | $ | (9,500 | ) | $ | (9,257 | ) | $ | (9,020 | ) | $ | (6,113 | ) | $ | (4,478 | ) | |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||
After-Tax Cash Flow |
| $ | 250 |
| $ | 244 |
| $ | 237 |
| $ | 231 |
| $ | 157 |
| $ | 115 |
| |
Present Value Factor |
| 0.930 |
| 0.802 |
| 0.692 |
| 0.596 |
| 0.064 |
| 0.028 |
| |||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||
Present Value of Cash Flows |
| $ | 233 |
| $ | 195 |
| $ | 164 |
| $ | 138 |
| $ | 10 |
| $ | 3 |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||
Total PV of Cash Flows (FV of Asset) |
| $ | 1,465 |
|
|
|
|
|
|
|
|
|
|
| ||||||
In this example, the acquired customer revenue is attrited at a 4% annual rate by multiplying the prior year factor by 96%. As mentioned above, both the fair value of the customer relationships related to the Amrep and Waterbury acquisitions and those assets’ useful lives were derived using the excess earnings method under the income approach. In this example (which we simplified for illustrative purposes), the sum of discounted or present valued cash flows would equal the fair value of the assets for which the income approach was conducted. The economic benefit generated by sales from repeat customers continues through Year 30. However, substantially all, or approximately 95% of that benefit, is realized by Year 18. Therefore, using the above example and in conjunction with consideration of other pertinent factors discussed further in our below-listed response to Comment No. 6, we would assign a useful life of 18 years to this asset.
In addition to analyzing years of historical sales data from acquired customers, we also advise the Staff that during due diligence we made inquires of the parties from which we acquired Amrep and Waterbury, as well as their management teams. Responses to those inquiries support the customer relationships’ useful lives derived from these
models. Further, after analyzing the years-long historical sales data and performing these inquiries, we noted our own historical experience with similar customers corroborates our useful life determinations. We also considered additional factors that are included in our below-listed responses.
4. In arriving at your attrition percentages, you divided the aggregate sales generated by lost customers by total sales. Please tell us what consideration you gave to also evaluating the actual number of customers lost by the total number of customers acquired in your analysis.
RESPONSE: We believe the benefit of acquired customer relationships rests with the value of future sales attributable to those relationships rather than the number of customers acquired that, without sales, do not possess a recognizable benefit. For this reason, we based the fair value determination of our acquired customer relationships and underlying assumptions on sales rather than customer count.
Our third party valuation professionals advised us that an attrition rate derived from customer count is not a generally accepted valuation practice except in rare instances when sales are evenly distributed among individual customers throughout a customer base. In this scenario, each individual customer’s contribution to total sales is equal and proportional. Additionally, we were advised that customer count-based attrition calculations may be used to compensate for an inability to obtain reliable historical sales data.
We advise the Staff that we obtained reliable historical sales data from the parties from which we acquired Waterbury and Amrep prior to commencing our valuation work. We also advise the Staff that among the hundreds of customers acquired, each individual customer’s contribution to total sales differs significantly. We believe all of these reasons support an attrition rate calculation based on sales to customers rather than customer count.
5. You have been doing business with retailers such as The Home Depot, True-Value, and Ace Hardware for approximately 30 years. The customer relationships at the acquisition dates had already existed for a long period of time (certain for more than 25 years). While it would appear that you would have long-lasting relationships with certain customers, please help us understand your basis for assuming that this would be the case with nearly all of the customers acquired.
RESPONSE: Amrep and Waterbury have customer relationships with leaders in the product categories served by those entities. In our previous response, we confidentially provided the Staff with sales data for Amrep’s Professional Product Group product line. Customers served by this product family include janitorial and sanitation distributors, master wholesalers, specialty industry distributors, and others. Leaders in these categories served by Amrep include Lasgasse-Sweet, Fastenal, Standardized Sanitation Systems (Triple S), Aramsco, and Office Depot. Information previously supplied to the Staff also referenced Amrep’s ALPMG product line. Customers served by this product family include Asian automotive original equipment manufacturers that source brake and carburetor cleaners, automatic transmission fluids, coolants, and other related products. Leaders in these categories served by Amrep include Toyota Motor Sales USA, Nissan North America, Kia, Subaru, etc. Waterbury has similarly enjoyed relationships with large retailers and distributors.
In its previous comment letter, the Staff requested to be informed of “what consideration [we] gave to using [our] own historical experience with similar customers, if any, over a longer period of time in evaluating the appropriateness of these amortization periods”. We also have lengthy historical experience with retail, distribution, and national account customers that are leaders in their end markets of which The Home Depot, True-Value, and Ace Hardware are representative, recognizable examples. Our response to the Staff confirmed that we did consider this experience when evaluating the useful lives of acquired relationships.
However, in answering the Staff’s request for information regarding our own historical experience, we did not intend to imply that our own experience singularly formed the basis for the useful lives of all acquired customer relationships. Rather, our own historical experience merely corroborates the output of extensive useful life computations of the acquired customer relationships, the information obtained from the above-mentioned inquires, and the other procedures referenced throughout this correspondence.
6. In regards to attaining greater benefit from an acquired customer base in earlier versus later periods after acquisition, you address how you determined that an accelerated method of amortization should not be used. However, it is not clear what consideration you gave as to whether this factor could lead to a shorter period of amortization. Please advise.
RESPONSE: We respectfully refer the Staff to our above-listed responses to Comment Nos. 3, 4, and 5 where we provide detailed information regarding how we determined the useful lives of acquired customer relationships.
The following is a summary of our evaluation of the useful lives and amortization method for acquired customer relationships:
· We considered that these identifiable intangibles have a finite useful life based on historical and projected customer attrition.
· We determined that the useful life for each product-line specific customer relationship was the period that the asset is expected to contribute directly or indirectly to our future cash flows consistent with ASC 350-30-35-2. This useful life is consistent with the point in time at which we expect to realize 95% of the projected future cash flows of each relationship. We also believe this is consistent with ASC 350-30-35-3, which states, in part, “…if an income approach is used to measure the fair value of an intangible asset, in determining the useful life of the intangible asset for amortization purposes, an entity shall consider the period of expected cash flows used to measure the fair value of the intangible asset….”
· We considered the factors in ASC 350-30-35-3 (a.) through (f.), including whether (1) there are legal, regulatory, contractual, competitive, economic or other factors that would limit the useful lives, other than the impact of competition already inherently included in the determination of attrition rates, and (2) the useful lives determined for the acquired assets are consistent with historical experience. We advise the Staff that there are no legal or contractual restrictions on our ability to continue our relationships with the acquired customers. We do not operate in a highly regulated industry and there are no regulatory restrictions on our ability to continue relationships with the acquired customers. We operate in a competitive industry, but we know of no specific competitive or economic factors that would affect our useful life
determinations. We expect the sales orders and patterns from the acquired customers to be consistent with our own historical experience with similar customers. In addition, we specifically considered the “churn” rate of customers through our above-described customer attrition analyses.
· As noted in our response to Comment No. 3, the excess earnings method under the income approach calculates the economic benefit associated with our acquired customer relationships to continue well beyond the point at which our useful lives are based (in that example, we expect 30 years of benefit and assigned a useful life of 18 years).
· We also considered that the annual amortization expense recognized for these intangibles is not highly sensitive to changes in our useful life assumptions. For example, the incremental expense caused by a 20% or almost 4-year reduction in our weighted average useful life assumptions would result in less than a 2% reduction in our estimated pre-tax income for fiscal year 2011. Based on our consideration of the factors described throughout this response, we concluded that an even shorter amortization period was not appropriate, and that we have properly considered all pertinent information available when estimating the useful lives of acquired customer relationships.
· The valuation model projections used to value the customer relationships allow us to conclude a reasonable and reliable overall fair value and expected useful life for acquired customer relationships. However, they do not allow for a reliably determinable pattern of economic benefit that could be used to derive an amortization period other than straight-line. Therefore, we used a straight-line amortization method in accordance with ASC 350-30-35-6. In concluding on the appropriateness of this method, we also concluded that the annual differences between straight-line and attrition-based amortization methods were not material (e.g. less than 1% of pre-tax income for fiscal 2011).
We further advise the Staff that we have internal controls that enable us to monitor continuously our customer relationships for the loss of significant customers or customer groupings that would warrant an interim impairment measurement. We have not experienced the loss of significant customers or customer groupings for either Amrep or Waterbury subsequent to their acquisition.
Form 10-Q for the Period Ended May 31, 2011
General
7. Please address the above comments in your interim filings as well, as applicable.
RESPONSE: We will incorporate the above responses in our interim filings, as applicable.
If the Staff has any questions or comments, we encourage the undersigned be contacted at (404) 603-7815.
| Sincerely, |
|
|
| /s/ Mark R. Bachmann |
|
|
| Mark R. Bachmann |
| Executive Vice President and Chief Financial Officer |
cc: Ms. Nudrat Salik, Staff Accountant, United States Securities and Exchange Commission
Mr. John K. Morgan, Zep Inc. Chairman, President and Chief Executive Officer
Mr. Philip A. Theodore, Zep Inc. Vice President, General Counsel and Corporate Secretary
Mr. Beau B. Bradley, Ernst & Young LLP
Mr. Christopher Green, Hunton & Williams LLP
