WESTINGHOUSE AIR BRAKE TECHNOLOGIES CORPORATION.

1001 Air Brake Avenue

Wilmerding, PA 15148

September 16, 2011

Via EDGAR

Lyn Shenk

Branch Chief

United States Securities and Exchange Commission

100 F Street, NE

Washington, DC 20549

 

RE:

Westinghouse Air Brake Technologies Corporation

Form 10-K for the Year Ended December 31, 2010

Filed February 25, 2011

File No. 033-90866

Dear Mr. Shenk:

We have reviewed the comments and recommendations of the Staff set forth in your letter dated August 23, 2011. On behalf of Westinghouse Air Brake Technologies Corporation, this letter will set forth our responses to the Staff’s comments. Note that all page numbers cited in our responses below refer to the page numbers in the Form 10-K filed with the SEC on February 25, 2011.

Form 10-K for Fiscal Year Ended December 31, 2010

Management’s Discussion and Analysis

Results of Operations, page 25

 

1.	Please revise to discuss and analyze the reasons underlying non acquisition-related changes in sales. For example, explain how and why specialty product sales in the freight group increased $63.2 million from 2009 to 2010.

---

Lyn Shenk

September 16, 2011

Page 2

 

Future filings will be revised to disclose the reasons underlying non acquisition-related changes in sales. The following is an example of the revisions as applied to the 10-K for the fiscal year ended December 31, 2010:

Net sales increased by $105.4 million to $1,507.0 million in 2010 from $1,401.6 million in 2009. The increase is due to higher sales of $100.2 million from increased customer demand for freight original equipment and aftermarket products, freight overhaul and remanufacturing services, and sales related to acquisitions of $86.1 million. Partially offsetting this increase was lower sales of $85.1 million, resulting from the completion of a certain transit original equipment contract, lower demand for transit car overhauls and the completion of certain transit locomotive build contracts. The Company realized a net sales increase of $4.2 million due to favorable effects of foreign exchange, but net earnings were not materially impacted by foreign exchange. Net income for 2010 was $123.1 million or $2.56 per diluted share. Net income for 2009 was $115.1 million or $2.39 per diluted share. Net income increased due to higher sales volume and operating margins, partially offset by higher income tax expense.

Freight Group sales increased by $196.1 million or 33.3% due to higher sales of $81.8 million from acquisitions, $63.2 million for specialty products, primarily resulting from increased customer demand for original equipment heat exchange products and aftermarket rail products, $20.4 million for brake products, resulting from increased rail traffic and aftermarket demand for brakes and valves, $13.6 million for other freight products, primarily resulting from increased international demand, and freight overhaul and remanufacturing services increased $3.0 million. For the Freight Group, net sales improved by $14.1 million due to the favorable effects of foreign exchange.

Transit Group’s sales decreased by $90.7 million or 11.2%. Sales decreased $65.2 million because of the completion of a major original equipment contract early in 2010. Sales also decreased by $19.0 million due to lower levels of transit car overhauls and the completion of certain contracts for the manufacture of locomotives in 2010. These decreases were partially offset by sales from acquisitions of $4.3 million. For the Transit Group, net sales were reduced by $9.9 million due to unfavorable effects of foreign exchange.

 

2.

We note that cost of sales is material to your consolidated results but you do not provide a discussion and analysis of these costs. We believe you should revise your disclosure to provide a direct comparative discussion and analysis of cost of sales on a consolidated basis at an appropriate level of detail. Your disclosure should quantify and analyze the impact of each (and not netted) significant component of cost of sales that caused cost of sales to vary materially (or not vary when expected to) between comparative periods, with explanation of the associated underlying reasons. We also believe that these disclosures, in particular in regard to underlying reasons for material changes in costs of sales, are appropriate at the segment level when a change in a cost of sales of a segment materially impacts the segment’s measure of profit or performance. In this regard, we believe you should quantify cost of sales associated with each segment to enable investors to understand their magnitude and relative impact on each segment’s results. Please revise your disclosure accordingly and provide us with a copy of your intended revised disclosure.

---

Lyn Shenk

September 16, 2011

Page 3

 

Future filings will be revised to disclose a discussion and analysis for cost of sales. The following is an example of the revisions as applied to the 10-K for the fiscal year ended December 31, 2010:

Cost of Sales and Gross profit Cost of Sales increased by $49.6 million to $1,057.9 million in 2010 compared to $1,008.3 million in 2009. In 2010, cost of sales, as a percentage of sales was 70.2% compared to 71.9% in 2009. This decrease is the result of (i) changes in product mix in which higher margin product sales (freight and aftermarket) increased as a percentage of total sales compared to other products, (ii) realized cost savings from downsizing and consolidation actions, and (iii) decreased costs in the transit segment due to efficiencies realized on certain long-term contracts.

During 2010, raw material costs decreased as a percentage of sales from approximately 43% in 2009 to 41% in 2010. Labor costs as a percentage of sales were approximately 11% in 2009 and 2010. Overhead costs as a percentage of sales were approximately 18% in 2009 and 2010. Freight group raw material costs decreased as a percentage of sales from approximately 43% in 2009 to 40% in 2010. Freight group labor costs as a percentage of sales were approximately 11% in 2009 and 2010, and overhead costs decreased as a percentage of sales from approximately 20% in 2009 to 18% in 2010. Transit group raw material costs decreased as a percentage of sales from approximately 43% in 2009 to 42% in 2010. Transit group labor costs as a percentage of sales were approximately 10% in 2009 and 2010, and overhead costs increased as a percentage of sales from approximately 14% in 2009 to 16% in 2010. In general, raw material costs as a percentage of sales decreased reflecting the higher mix of revenue generated from contracts delivering train control and other related services, which carries a higher labor component as cost of sales. Overhead costs vary as a percentage of sales depending on product mix and changes in sales volume. In addition, included in costs of sales is warranty expense. The provision for warranty expense is generally established for specific losses, along with historical estimates of customer claims as a percentage of sales, which can cause variability in warranty expense between quarters. Warranty expense was $2.4 million higher in 2010 compared to 2009 due to increased Freight Group sales.

Gross profit increased to $449.1 million in 2010 compared to $393.3 million in 2009, for the reasons discussed above. Accordingly, in 2010, gross profit, as a percentage of sales, was 29.8% compared to 28.1% in 2009.

Operating Expenses, pages 27 and 28

 

3.

Please quantify each factor cited in explaining variances so that investors may understand its relative impact. For example, several factors are cited in regard to the variances in selling, general and administrative expenses between the comparative periods reported without quantification. Refer to section 501.04 of the Codification of Financial Reporting Releases for guidance.

---

Lyn Shenk

September 16, 2011

Page 4

 

Future filings will be revised to quantify each factor cited in explaining variances. The following is an example of the revisions as applied to the 10-K for the fiscal year ended December 31, 2010:

Operating expenses The following table shows our operating expenses:

 

		For the year ended December 31,
In thousands		2010				2009				Percent Change
Selling, general and administrative expenses		$	195,892			$	160,998				21.7	%
Engineering expenses			40,203				42,447				(5.3	)%
Amortization expense			10,173				9,849				3.3	%
Total operating expenses		$	246,268			$	213,294				15.5	%

Selling, general, and administrative expenses increased $34.9 million in 2010 compared to 2009 primarily due to $17.3 million of expenses from acquisitions, and $16.7 million of incentive and non-cash compensation. Engineering expense decreased by $2.2 million in 2010 compared to 2009 as the company focused engineering resources on completing original equipment contracts. Costs related to engineering for specific customer contracts are included in the labor element of cost of sales. Total operating expenses were 16.3% and 15.2% of sales for 2010 and 2009, respectively.

Critical Accounting Policies, page 35

 

4.

We believe your disclosure could provide greater insight into the quality, sensitivity and variability regarding the factors that have or may materially affect amounts associated with the policies you have indicated. Your disclosure should be explicit as to which of the identified factors are most sensitive to change, deviations of estimates and assumptions from actual results, and the circumstances that resulted in revised assumptions in the past or that could lead to material changes in the future. To the extent practicable and material, you should provide quantitative disclosure, with an analysis of how actual results may differ from your estimates under different assumptions and conditions that you have considered. Refer to Section V of “Interpretation: Commission Guidance Regarding Management’s Discussion and Analysis of Financial Condition and Results of Operations” available on our website at http://www.sec.gov/rules/interp/33-8350.htm for further guidance. Please revise your disclosure as appropriate.

---

Lyn Shenk

September 16, 2011

Page 5

 

Upon review of our Critical Accounting Policies, we believe the guidance is directly applicable to our policies for Goodwill and Indefinite-Lived Intangibles and Accounting for Pensions and Postretirement Benefits. Future filings will be revised to provide additional qualitative and quantitative disclosures. The following is an example of the revisions as applied to the 10-K for the fiscal year ended December 31, 2010:

 

Description		Judgments and Uncertainties		Effect if Actual Results Differ From Assumptions
Goodwill and Indefinite-Lived Intangibles:
Goodwill and indefinite-lived intangibles are required to be tested for impairment at least annually. The Company performs its annual impairment test during the fourth quarter and more frequently when indicators of impairment are present. The Company reviews goodwill for impairment at the reporting unit level. The evaluation of impairment involves comparing the current fair value of the business to the recorded value (including goodwill).		We use a combination of a guideline public company market approach and a discounted cash flow model (“DCF model”) to determine the current fair value of the business. A number of significant assumptions and estimates are involved in the application of the DCF model to forecast operating cash flows, including markets and market share, sales volume and pricing, costs to produce and working capital changes. During the fourth quarter of 2010, the Company used a discount rate for each of it reporting units ranging from 11% to 13% and a terminal growth of 1% to 3%. The difference in discount rates is based on the underlying markets and risks associated with each reporting unit.		Management considers historical experience and all available information at the time the fair values of its business are estimated. However, actual amounts realized may differ from those used to evaluate the impairment of goodwill.   If actual results are not consistent with our assumptions and judgments used in estimating future cash flows and asset fair values, we may be exposed to impairment losses that could be material to our results of operations. Testing goodwill for impairment requires us to estimate fair values of reporting units using significant estimates and judgmental factors. The key estimates and factors used in our discounted cash flow valuation include revenue growth rates and profit margins based on internal forecasts, terminal value, and the weighted-average cost of capital used to discount future cash flows. The compound annual growth rate for revenue during the first five years of our projections was approximately 12.6%. The terminal value was calculated assuming projected growth rates of 1.0% to 3.0% depending on the respective unit’s growth outlook. This rate reflects our estimate of long-term growth into perpetuity and approximates the long-term gross domestic product growth expected on a global basis. Operating profit margins were projected to return to historical norms during fiscal 2012 and fiscal 2013 in the individual reporting units with the exception of the freight group which expects strong margin growth as product demand increases over the projection period. The estimated weighted-average cost of capital for the reporting units was determined to range from 10.6% to 12.6% depending on each unit’s business risk based upon an analysis of similar companies and their debt to equity mix, their related volatility and the size of their market capitalization. We also consider any

---

Lyn Shenk

September 16, 2011

Page 6

 

Description		Judgments and Uncertainties		Effect if Actual Results Differ From Assumptions
				additional risk of each individual reporting unit achieving its forecasts, and adjusted the weighted-average cost of capital applied when determining each reporting unit’s estimated fair value. Future changes in these estimates and assumptions could materially affect the results of our goodwill impairment tests. For example, a decline in the terminal growth rate greater than 50 basis points would decrease fair market value by $133.3 million, or an increase in the weighted-average cost of capital by 100 basis points would result in a decrease in fair market value by $373.4 million. Even with such changes the fair value of the reporting units would be greater than their net book values as of our valuation date, necessitating no Step 2 calculations.
Accounting for Pensions and Postretirement Benefits:
These amounts are determined using actuarial methodologies and incorporate significant assumptions, including the rate used to discount the future estimated liability, the long-term rate of return on plan assets and several assumptions relating to the employee workforce (salary increases, medical costs, retirement age and mortality).		Significant judgments and estimates are used in determining the liabilities and expenses for pensions and other postretirement benefits.   The rate used to discount future estimated liabilities is determined considering the rates available at year-end on debt instruments that could be used to settle the obligations of the plan. The long-term rate of return is estimated by considering historical returns and expected returns on current and projected asset allocations and is generally applied to a five-year average market value of assets.		If assumptions used in determining the pension and other postretirement benefits change significantly, these costs can fluctuate materially from period to period. The key assumptions in determining the pension and other postretirement expense and obligation include the discount rate, expected return on assets and health care cost trend rate. For example, a 1% decrease or increase in the discount rate used in determining the pension and postretirement expense would increase expense $2.2 million or decrease expense $2.0 million, respectively. A 1% decrease or increase in the discount rate used in determining the pension and postretirement obligation would increase the obligation $32.1 million or decrease the obligation $28.0 million, respectively. A 1% decrease or increase in the expected return on assets used in determining the pension expense would

---

Lyn Shenk

September 16, 2011

Page 7

 

Description		Judgments and Uncertainties		Effect if Actual Results Differ From Assumptions
				increase expense $1.7 million or decrease expense $1.7 million, respectively. A 1% decrease or increase in the health care cost trend rate used in determining the postretirement expense would decrease expense $0.4 million or increase expense $0.4 million, respectively. A 1% decrease or increase in the health care cost trend rate used in determining the postretirement obligation would decrease the obligation $3.6 million or increase the obligation $4.2 million, respectively.

******

If you have any further questions or comments, please contact the undersigned at (412) 825-1317 or contact our counsel at K&L Gates LLP, Dave DeNinno (412-355-6513).

 

 

cc:

P. Dugan

D. Labate

D. DeNinno, Esq.