January 11, 2010
Lyn Shenk, Branch Chief
Division of Corporation Finance
U.S. Securities and Exchange Commission
100 F Street, N.E.
Washington, D.C. 20549
| Re: | Union Pacific Corporation |
File No. 001-06075
Form 10-K for the fiscal year ended December 31, 2008
Form 10-Q for the quarterly period ended March 31, 2009
Dear Mr. Shenk:
This letter is in response to the comment letter, dated October 30, 2009, addressed to Mr. Robert M. Knight, Jr., Executive Vice President-Finance and Chief Financial Officer of Union Pacific Corporation (the “Company”), regarding the comments of the Staff of the Securities and Exchange Commission with respect to the Company’s September 29, 2009 correspondence. In addition, this letter reflects responses to various telephonic discussions with the Staff.
For the convenience of the Commission Staff, we reproduce the text of each numbered paragraph in the comment letter and follow with our responses.
We respectfully submit the following information and comments with respect to each comment contained in the comment letter.
* * * * *
Form 10-K for the Fiscal Year Ended December 31, 2008
Item 7. Management’s Discussion and Analysis of Financial Condition and Results of Operations
Liquidity and Capital Resources
Investing Activities, page 38
| 1. | Refer to your response to our prior comment number 4. You state that the table on page 38 of your Form 10-K for the fiscal year ended December 31, 2008 is designed to show how you manage your capital programs. However, we believe that your table would be more meaningful to the readers of your financial statements if it was supplemented by |
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 2
| additional tabular disclosure and/or a narrative discussion of (i) the number of track miles of rail replaced, (ii) the number of crossties replaced, (iii) the number of new track miles of rail installed, (iv) the number of new crossties installed, (v) the number of track miles of rail resurfaced, and (vi) any additional information that would allow readers to further analyze the factors contributing to the fluctuations in your capital expenditures. In this regard, also consider expanding the tabular disclosure currently presented on page 38 to separately quantify your capital expenditures related to the replacement of rail, the replacement of ties, resurfacing activities, and any other significant activities related to the replacement or renewal of your existing depreciable road assets and/or infrastructure. Please provide your proposed expanded disclosure as part of your response. |
RESPONSE:
We note your comment and will revise our future filings on Form 10-K to provide additional detail regarding our cash capital expenditures. Although you requested that we report both the number of crossties replaced and the number of new crossties installed, we propose reporting only the number of new crossties installed because this number includes crosstie replacements. The tables presented below and on the following page will replace the table that was previously included in the Liquidity and Capital Resources – Investing Activities section of the MD&A of our Form 10-K filings.
The tables below detail cash capital investments (including capital leases) and track statistics for the years ended December 31, 2008, 2007, and 2006:
| Millions of Dollars | 2008 | 2007 | 2006 | ||||||
| Rail |
$ | 646 | $ | 628 | $ | 574 | |||
| Ties |
425 | 404 | 398 | ||||||
| Ballast |
243 | 206 | 183 | ||||||
| Other [a] |
386 | 355 | 332 | ||||||
| Total road infrastructure replacements |
1,700 | 1,593 | 1,487 | ||||||
| Line expansion and other capacity projects |
488 | 419 | 350 | ||||||
| Commercial facilities |
254 | 115 | 160 | ||||||
| Total capacity projects and commercial facilities |
742 | 534 | 510 | ||||||
| Locomotives and freight cars |
164 | 263 | 135 | ||||||
| Technology and other |
174 | 106 | 110 | ||||||
| Total cash capital investments |
$ | 2,780 | $ | 2,496 | $ | 2,242 | |||
| [a] Other includes bridges and tunnels, signals, and other road assets |
|||||||||
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 3
| 2008 | 2007 | 2006 | ||||
| Track miles of rail replaced |
810 | 877 | 866 | |||
| Track miles of rail capacity expansion |
118 | 79 | 127 | |||
| New crossties installed (thousands) |
4,599 | 4,267 | 4,573 | |||
| Miles of track surfaced |
11,369 | 12,495 | 8,239 |
Critical Accounting Policies
Property and Depreciation, page 49
| 2. | Please refer to your response to our prior comment number 3. We note that you do not recognize a gain or loss upon the normal retirement (or replacement) of depreciable rail property accounted for pursuant to the group method of depreciation, as you assume that, on average, such assets are fully depreciated at the time they are retired. However, we note that your response does not (i) explain how you distinguish between normal and abnormal retirements of your depreciable rail assets or (ii) discuss your accounting treatment for abnormal retirements. In this regard, while it appears reasonable for you to characterize the replacements of depreciable rail assets that have reached or exceeded their expected service lives as normal retirements, it is not clear from your response or your disclosure how you determine the appropriate characterization for replacements of depreciable rail assets that occur prior to the end of such assets’ expected service lives. More specifically, it appears that circumstances could exist for which it would be appropriate to evaluate whether the replacement of depreciable rail assets prior to the end of their expected service lives reflects or suggests an abnormal retirement. Given that it is inherent under group depreciation that a significant portion of your depreciable rail assets will not reach their expected service lives prior to retirement (e.g., per your response, 45 percent of the depreciable rail assets that you retired in fiscal year 2008 did not reach their expected service lives), we believe that it is important for you to disclose how you evaluate whether premature asset replacements should be characterized as normal or abnormal retirements. Furthermore, we believe that your accounting policy regarding property and depreciation should specifically address the accounting treatment applied to abnormal retirements. Please expand the disclosure in the “Critical Accounting Policies” section of MD&A and the footnotes to your financial statements, as appropriate. In addition, please provide your proposed expanded disclosure as part of your response. |
RESPONSE:
In the Overview of Group Depreciation provided in our response to your August 31, 2009 comment letter, we noted that the calculation of depreciation expense for large groups of assets requires the use of averages, in particular for service life. Average values are required because not all of the assets in the groups of similar function/nature experience the same service life.
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 4
That is, despite the fact that the assets in the group are homogeneous, they experience lives that are dispersed over a range.1 This range can be wide for multiple reasons including physical factors (e.g., wear and tear, decay or deterioration), functional factors (e.g., obsolescence, changes in technology, changes in demand, or requirements of public authorities), and contingent factors (e.g., accidents/casualties or severe weather conditions). One of the major advantages of the group depreciation method is that it deals effectively with dispersions in asset lives, which are the focus of periodic depreciation studies. The life studies use actuarially-based methods to analyze asset retirements and, in turn, to compute the average life of the property (i.e., the estimated service life). Because the estimated service life represents the mean, it should be expected that 50 percent of the assets in a group are retired prior to reaching their estimated service life and 50 percent retired after. We previously indicated that 45 percent of the depreciable rail assets that we retired in fiscal year 2008 did not reach their expected service lives. This does not imply that we may have had “premature asset replacements”, but rather that our retirement experience may indicate that a change in the estimated useful life was warranted.
Although the group method of depreciation does not recognize gains or losses on normal retirements or replacements of depreciable property, the group method provides for adjustments to earnings in the event of abnormal retirements. The 2007 PricewaterhouseCoopers accounting guide for the rail industry states that, “These occurrences, which result in a gain or loss, have been relatively rare historically and usually include significant line and equipment sales.” With respect to our policy, we define a retirement as abnormal if it meets each of the following three conditions: (i) is unusual, (ii) is material in amount, and (iii) varies significantly from the retirement profile identified through our depreciation studies. When assessing materiality, we consider all relevant facts and circumstances, including the significance of a retirement to our financial statements, in making our determination.
We note your comment and propose including the following in the Critical Accounting Policies section of the MD&A of future Form 10-K filings:
Property and Depreciation – Our railroad operations are highly capital intensive and our large base of homogeneous, network-type assets turns over on a continuous basis. Each year we develop a capital program for the replacement of assets and for the acquisition or construction of assets that enable us to enhance our operations or provide new service offerings to customers. Assets purchased or constructed throughout the year are capitalized if they meet applicable minimum units of property criteria. Properties and equipment are carried at cost and are depreciated on a straight-line basis over their estimated service lives, which are measured in years except for rail in high-density traffic corridors (i.e., all rail lines except for those subject to abandonment, yard and switching tracks, and electronic yards), which are measured in millions of gross tons per mile of
| 1 | William M. Stout, P.E., A Comparison of Component and Group Depreciation For Large Homogeneous Groups of Network Assets, a presentation to the Accounting Standards Executive Committee of the American Institute of Certified Public Accountants |
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 5
track. We use the group method of depreciation in which all items with similar characteristics, use, and expected life are grouped together in asset classes, and are depreciated using composite depreciation rates. The group method of depreciation treats each asset class as a pool of resources, not as singular items. We currently have more than 60 depreciable asset classes, and we may increase or decrease the number of asset classes due to changes in technology, asset strategies, or other factors.
We determine the estimated service lives of depreciable railroad property by means of depreciation studies. We perform depreciation studies at least every three years for equipment and every six years for track assets (i.e., rail and other track material, ties, and ballast) and other roadway property. Our depreciation studies take into account the following factors:
| • | statistical analysis of historical patterns of use and retirements of each of our asset classes; |
| • | evaluation of any expected changes in current operations and the outlook for continued use of the assets; |
| • | evaluation of technological advances and changes to maintenance practices; and |
| • | expected salvage to be received upon retirement. |
For rail in high-density traffic corridors, we measure estimated service lives in millions of gross tons per mile of track. It has been our experience that the lives of rail in high-density traffic corridors are closely correlated to usage (i.e., the amount of weight carried over the rail). The service lives also vary based on rail weight, rail condition, (e.g., new or secondhand), and rail type (e.g., straight or curve). Our depreciation studies for rail in high density traffic corridors consider each of these factors in determining the estimated service lives. For rail in high-density traffic corridors, we calculate depreciation rates annually by dividing the number of gross ton-miles carried over the rail (i.e., the weight of loaded and empty freight cars, locomotives and maintenance of way equipment transported over the rail) by the estimated service lives of the rail measured in millions of gross tons per mile. Rail in high-density traffic corridors accounts for approximately 70 percent of the historical cost of rail and other track material. Based on the number of gross ton-miles carried over our rail in high density traffic corridors during 2009, the estimated service lives of the majority of this rail ranged from 14 years to 30 years. For all other depreciable assets, we compute depreciation based on the estimated service lives of our assets as determined from the analysis of our depreciation studies. Changes in the estimated service lives of our assets and their related depreciation rates are implemented prospectively.
Estimated service lives of depreciable railroad property may vary over time due to changes in physical use, technology, asset strategies, and other factors that will have an impact on the retirement profiles of our assets. We are not aware of any specific factors that are reasonably likely to significantly change the estimated service lives of our assets.
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 6
Actual use and retirement of our assets may vary from our current estimates, which would impact the amount of depreciation expense recognized in future periods.
Changes in estimated useful lives of our assets due to the results of our depreciation studies could significantly impact future periods’ depreciation expense and have a material impact on our Consolidated Financial Statements. If the estimated useful lives of all depreciable assets were increased by one year, annual depreciation expense would decrease by approximately $48 million. If the estimated useful lives of all depreciable assets were decreased by one year, annual depreciation expense would increase by approximately $52 million. Our recent depreciation studies have resulted in changes in depreciation rates for some asset classes, which did not significantly affect our annual depreciation expense.
Under group depreciation, the historical cost (net of salvage) of depreciable property that is retired or replaced in the ordinary course of business is charged to accumulated depreciation and no gain or loss is recognized. The historical cost of certain track assets is estimated using (i) inflation indices published by the Bureau of Labor Statistics and (ii) the estimated useful life of the assets as determined by our depreciation studies. The indices were selected because they closely correlate with the major costs of the properties comprising the applicable track asset classes. Because of the number of estimates inherent in the depreciation and retirement processes and because it is impossible to precisely estimate each of these variables until a group of property is completely retired, we continually monitor the estimated service lives of our assets and the accumulated depreciation associated with each asset class to ensure our depreciation rates are appropriate.
For retirements of depreciable railroad properties that do not occur in the normal course of business, a gain or loss may be recognized if the retirement meets each of the following three conditions: (i) is unusual, (ii) is material in amount, and (iii) varies significantly from the retirement profile identified through our depreciation studies. During the last three fiscal years, no gains or losses were recognized due to the retirement of depreciable railroad properties. A gain or loss is recognized in other income when we sell land or dispose of assets that are not part of our railroad operations.
We note your comment and propose including the following in the Properties footnote of future Form 10-K filings. Although not shown below, we will continue to include the table listing the major categories of property and equipment, as well as the weighted-average composite depreciation rate for each category, in future Form 10-K and Form 10-Q filings. In addition, pursuant to our discussion on January 5, 2010, we will amend the properties table to include accumulated depreciation by major category of property and equipment.
Property and Depreciation – Our railroad operations are highly capital intensive and our large base of homogeneous, network-type assets turns over on a continuous basis. Each year we develop a capital program for the replacement of assets and for the acquisition or
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 7
construction of assets that enable us to enhance our operations or provide new service offerings to customers. Assets purchased or constructed throughout the year are capitalized if they meet applicable minimum units of property criteria. Properties and equipment are carried at cost and are depreciated on a straight-line basis over their estimated service lives, which are measured in years except for rail in high-density traffic corridors (i.e., all rail lines except for those subject to abandonment, yard and switching tracks, and electronic yards), which are measured in millions of gross tons per mile of track. We use the group method of depreciation in which all items with similar characteristics, use, and expected life are grouped together in asset classes, and are depreciated using composite depreciation rates. The group method of depreciation treats each asset class as a pool of resources, not as singular items. We currently have more than 60 depreciable asset classes, and we may increase or decrease the number of asset classes due to changes in technology, asset strategies, or other factors.
We determine the estimated service lives of depreciable railroad assets by means of depreciation studies. We perform depreciation studies at least every three years for equipment and every six years for track assets (i.e., rail and other track material, ties, and ballast) and other roadway property. Our depreciation studies take into account the following factors:
| • | statistical analysis of historical patterns of use and retirements of each of our asset classes; |
| • | evaluation of any expected changes in current operations and the outlook for continued use of the assets; |
| • | evaluation of technological advances and changes to maintenance practices; and |
| • | expected salvage to be received upon retirement. |
For rail in high-density traffic corridors, we measure estimated service lives in millions of gross tons per mile of track. It has been our experience that the lives of rail in high-density traffic corridors are closely correlated to usage (i.e., the amount of weight carried over the rail). The service lives also vary based on rail weight, rail condition, (e.g., new or secondhand), and rail type (e.g., straight or curve). Our depreciation studies for rail in high density traffic corridors consider each of these factors in determining the estimated service lives. For rail in high-density traffic corridors, we calculate depreciation rates annually by dividing the number of gross ton-miles carried over the rail (i.e., the weight of loaded and empty freight cars, locomotives and maintenance of way equipment transported over the rail) by the estimated service lives of the rail measured in millions of gross tons per mile. For all other depreciable assets, we compute depreciation based on the estimated service lives of our assets as determined from the analysis of our depreciation studies. Changes in the estimated service lives of our assets and their related depreciation rates are implemented prospectively. Under group depreciation, the historical cost (net of salvage) of depreciable property that is retired or replaced in the ordinary course of business is charged to accumulated
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 8
depreciation and no gain or loss is recognized. The historical cost of certain track assets is estimated using (i) inflation indices published by the Bureau of Labor Statistics and (ii) the estimated useful life of the assets as determined by our depreciation studies. The indices were selected because they closely correlate with the major costs of the properties comprising the applicable track asset classes. Because of the number of estimates inherent in the depreciation and retirement processes and because it is impossible to precisely estimate each of these variables until a group of property is completely retired, we continually monitor the estimated service lives of our assets and the accumulated depreciation associated with each asset class to ensure our depreciation rates are appropriate.
For retirements of depreciable railroad properties that do not occur in the normal course of business, a gain or loss may be recognized if the retirement meets each of the following three conditions: (i) is unusual, (ii) is material in amount, and (iii) varies significantly from the retirement profile identified through our depreciation studies. A gain or loss is recognized in other income when we sell land or dispose of assets that are not part of our railroad operations.
When we purchase an asset, we capitalize all costs necessary to make the asset ready for its intended use. However, many of our assets are self-constructed. A large portion of our capital expenditures is for replacement of existing road infrastructure assets (program projects), which is typically performed by our employees, and for track line expansion (capacity projects). Costs that are directly attributable or overhead costs that relate directly to capital projects are capitalized. Direct costs that are capitalized as part of self-constructed assets include material, labor, and work equipment. Indirect costs are capitalized if they clearly relate to the construction of the asset. These costs are allocated using appropriate statistical bases.
General and administrative expenditures are expensed as incurred. Normal repairs and maintenance are also expensed as incurred, while costs incurred that extend the useful life of an asset, improve the safety of our operations or improve operating efficiency are capitalized.
Assets held under capital leases are recorded at the lower of the net present value of the minimum lease payments or the fair value of the leased asset at the inception of the lease. Amortization expense is computed using the straight-line method over the shorter of the estimated useful lives of the assets or the period of the related lease.
| 3. | Per your response to our prior comment number 3, you assume that, on average, your company’s rail property is fully depreciated upon normal retirement. We note that the validity of your assumption depends upon the accuracy of (i) your estimated service lives for each of your rail categories (i.e., measured in millions of ton-miles per mile) and (ii) your composite depreciation rates for your company’s other depreciable rail property – |
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 9
| both of which are determined based upon your depreciation studies. Given (a) the complexity of your depreciation studies, (b) the significant estimates, significant assumptions, and degree of judgment required to perform your depreciation studies, and (c) the impact of the results of your depreciation studies on your estimated asset lives and reported operating performance, we believe that your disclosure regarding “property and depreciation” in the “Critical Accounting Policies” section of MD&A should be expanded to include a detailed discussion of the material uncertainties associated with your depreciation studies. In this regard, please expand your disclosure to discuss information including, but not limited to, the following: |
| • | the manner in which your depreciation studies are performed; |
| • | factors that could cause variability in the results of your depreciation studies or the estimated asset service lives derived from your depreciation studies; |
| • | the nature of any material assumptions or estimates either included in your depreciation studies or required to derive the estimated asset service lives from your depreciation studies; |
| • | the sources of the information incorporated into your depreciation studies, including the underlying basis for any material assumptions and/or estimates; |
| • | the reasons why the estimates and assumptions incorporated in your depreciation studies, as well as the estimated depreciable lives derived from your depreciation studies, are subject to change; |
| • | the accuracy of your estimates of the service lives of your depreciable rail assets in the past – including a discussion and/or quantification of specific adjustments that you have made to your composite rates or estimated depreciable lives, the factors that you believe contributed in such adjustments, and the expected impact of current adjustments on your reported depreciation expense; and |
| • | whether the estimated service lives applied to your depreciable rail assets are reasonably likely to change in the future, including a discussion of any known factors that are expected to impact your estimates or the underlying assumptions. |
For further guidance, please refer to Section V of our interpretive release “Commission Guidance Regarding Management’s Discussion and Analysis of Financial Condition and Results of Operations,” which is available on our website at http://www.sec.gov/rules/interp/33-8350.htm. Please provide your proposed expanded disclosure as part of your response.
RESPONSE:
We note your comment and refer you to comment 2 above for our proposed disclosure regarding Property and Depreciation to be included in the Critical Accounting Policies section of the MD&A of future Form 10-K filings. In addition, please note the discussion of recent changes in our estimated service lives in the Results of Operations section of the MD&A of our 2009 filings on Form 10-Q.
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 10
| 4. | We have reviewed your response to our prior comment number 4, as well as the proposed disclosure provided as part of your response to our prior comment number 5. However, we do not believe that your proposed disclosure fully explains the significance of the assumptions and degree of uncertainty underlying your estimates of the historical installation costs of your replaced road assets. Based upon your response to our prior comment number 4, we note that it is impractical for you to track the original installation costs of the individual assets that comprise the asset groups that you use for depreciation purposes. As such, it appears that upon the retirement of road assets, you estimate those assets’ historical installation costs using information such as (i) the current replacement cost, (ii) the average age at which your assets are retired based upon your depreciation studies, and (iii) the Bureau of Labor Statistics’ inflation indices for the most significant asset components (e.g., steel and labor indices are used for rail). Given the nature of the assumptions and degree of judgment required to estimate the historical installation cost of road assets that are replaced, as well as the significance of your rail replacement activities, we believe that the uncertainties related to your estimates should be discussed in the “Critical Accounting Policies” section of your MD&A in significantly greater detail. In this regard, please expand your MD&A disclosure to provide a detailed discussion of the assumptions and variability attributable to your historical cost estimates. Please provide your proposed expanded disclosure as part of your response. |
RESPONSE:
We note your comment and refer you to comment 2 above for our proposed disclosure regarding Property and Depreciation to be included in the Critical Accounting Policies section of the MD&A of future Form 10-K filings.
Item 8. Financial Statements and Supplementary Data
Notes to the Consolidated Financial Statements
Note 9. Properties
Property and Depreciation, page 76
| 5. | For each of your last three fiscal years, please tell us the amount of costs capitalized for the replacement of depreciable rail property that was not contemplated in your original capital program for the respective fiscal year. |
RESPONSE:
During the years ended December 31, 2008, 2007, and 2006, we capitalized $77.6 million, $16.9 million, and $39.4 million for the replacement of depreciable road property that was not contemplated in our original capital programs for the respective fiscal years (these amounts are gross and do not reflect any insurance recoveries that may have been received). These additional expenditures were principally attributable to a 2008 mudslide that buried over
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 11
3,000 feet of our main line track near Eugene, Oregon and a 2005 storm that washed out several miles of track in the States of California and Nevada. Most of the road property destroyed by these two events had been in service for many years. As a result, the net book value of the assets retired was less than $10 million in each instance. We did not consider these retirements to be outside the normal course of business because (i) these events were not unusual inasmuch as we experience severe weather conditions every year, (ii) the net book value of the assets retired was not material, and (iii) the retirement of these assets did not vary significantly from the retirement profiles identified through our depreciation studies. Because we experience casualty events such as floods and mudslides every year, our annual capital plans consistently include more than $100 million for replacement of assets destroyed by these types of events. As we discussed during the conference call of January 5, 2010, we will continue to evaluate retirements not originally contemplated in our capital programs for appropriate accounting treatment in accordance with our policies, which would include consideration for disclosure in the MD&A and footnotes of future financial statements.
| 6. | We have reviewed your response to our prior comment number 5, as well as your proposed revisions to your footnote disclosure. However, we believe that your proposed disclosure should be expanded further to disclose the approximate percentage of your rail (i.e., based upon gross capitalized costs) that is located in high-density corridors, and therefore, depreciated pursuant to the unit of production method. In addition, we believe that you should further clarify for readers that the expected service life of your rail located in high-density corridors differs depending upon the category of rail, which is determined based upon rail weight, rail condition (e.g., new or secondhand), and rail characteristics (e.g., tangent or curve). Finally, if the expected service lives of your nine categories of rail located in high-density corridor differ significantly from one another, we believe that it may be meaningful to disclose the related range of expected service lives or composite depreciation rates, in addition to your current disclosure of the weighted-average composite rate for all rail. Please provide your proposed expanded disclosure as part of your response. |
RESPONSE:
We agree that it is meaningful to readers to include the approximate percentage of rail located in high-density traffic corridors and the basis of our depreciation studies for rail located in high-density traffic corridors. With respect to your comment about the expected service lives of our nine categories of rail located in high-density corridors, we have determined that two categories (new heavy-weight tangent and curve rail) account for over 85 percent of the cost of rail in Density 1, the most heavily used rail on our system. (We focused on Density 1 because it has a depreciation rate that is approximately three times that of Density 2 due to significantly higher usage.) Consequently, we propose including a range of the estimated service lives of these two categories of rail because they represent the majority of rail in high-density traffic corridors.
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 12
We note your comment and refer you to comment 2 above for our proposed disclosure regarding Property and Depreciation to be included in the Critical Accounting Policies section of the MD&A and in the Properties footnote of future Form 10-K filings.
| 7. | Refer to your response to our prior comment number 7. In your response, you quote the Accounting Research Manager’s guidance, which states that “costs should be capitalized when they appreciably extend the life . . . of the property, and should be expensed when they do not.” In addition, you define an extension of life as an expenditure that results in future economic benefits by enabling the use of an asset beyond its current estimated useful life. Based upon your definition of an extension of life, as well as your accounting policy, we do not concur with your conclusion that the costs attributable to rail grinding qualify for capitalization. We note from your response that you have capitalized rail grinding costs because you believe that the grinding is a replacement alternative, rather than the maintenance of depreciable rail assets. However, we note that rail grinding does not enable you to use your existing rail beyond its current estimated useful life, which would be the period over which you depreciate your rail. More specifically, you state in your response that rail grinding has been factored into the estimated service lives of your rail. As such, grinding appears to be a maintenance activity that is required for your rates of depreciation to be accurate. Therefore, the capitalization of rail grinding costs does not appear to comply with the criteria established by your own accounting policy. Furthermore, pursuant to footnote 10 to SFAS No. 142, factoring grinding into the estimated life of your rail would only appear appropriate if grinding were deemed to be a maintenance activity (i.e., not a capitalizable enhancement). |
In addition, you state that you capitalize costs
attributable to rail grinding because you believe that the grinding allows you to operate at higher speeds and maintain a more fluid system, which increases your capacity. In this regard, you have cited improvements in rail profile, rail surface,
and rail shape as examples of the direct benefits obtained from grinding. However, it appears that the benefits which you believe are achieved through rail grinding merely reflect the return of rail to its original condition, as opposed to
enhancements above the original capacity, safety, or functionality of your rail. We believe the accounting literature related to costs that extend the life, increase the capacity, or improve the safety or efficiency of property requires the
condition of the property to be improved after the costs are incurred as compared with the condition of that property when originally constructed or acquired. In this regard, refer to paragraph 1 of EITF 90-8. Also, refer to example 2 of Exhibit
90-8A to EITF 90-8, as the circumstances described in the example appear to be analogous to rail grinding. Based upon the information and accounting literature cited above, rail grinding appears to be a normal repair and maintenance activity.
Furthermore, it appears that the capitalization of rail grinding costs could result in the overstatement of depreciable rail assets on your balance sheet. In this regard, we note that you grind certain of your rail lines multiple times within an annual period. However,
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 13
the costs attributable to grinding are grouped with rail for depreciation purposes. As such, it appears that grinding costs are depreciated over a period of time that extends well beyond the period of associated benefits. In addition, based upon your response to our prior comment number 4, it does not appear that your accumulated grinding costs are contemplated in your estimates of the historical rail installation costs that should be charged to fixed assets and accumulated depreciation upon rail retirement. Therefore, it is unclear to us when and how grinding costs are appropriately retired.
Based upon the observations noted above, we believe that you should expense costs attributable to rail grinding as normal maintenance as incurred. We also believe that you should re-evaluate whether the costs associated with all other activities that are required to maintain the condition of your depreciable road assets through the end of their expected lives qualify for capitalization.
RESPONSE:
We appreciate your observations noted above; however, we believe that several pertinent factors indicate that rail grinding is, in fact, an improvement to the rail and that the costs of these improvements may be capitalized.
We perform grinding on new rail in high-density corridors shortly after it is installed (and occasionally before it is ever used), not to maintain this new rail, but to achieve the appropriate rail profile for traffic in these corridors and extend its useful life. Thus the asset in question, rail in high-density corridors, is in fact improved “from the condition of that property when originally acquired.” Analyses performed by a number of individuals of sections of new rail have shown demonstrable differences after grinding from the physical characteristics of the same sections of new rail that has not been subject to grinding. See Appendix A for a list of publications that provide support that rail grinding extends the life of rail (this list was previously provided as an attachment to our August 1, 2008 response letter). Subsequent grinding is performed on this rail to restore it to its improved condition as opposed to its condition upon acquisition. Therefore, rail grinding is an improvement to both new rail and rail in service. Clearly, this grinding is not returning the rail to its original condition, but improving the rail by modification of its profile and, in turn, lengthening its life.
Due to the amount of the cost (between $20,000 and $25,000 per day for each contracted grinding machine and its crew) and the impact grinding activities have on our network, we only grind a portion of our rail. During 2007 and 2008, less than half of our rail in high-density traffic corridors was ground. We could grind a greater percentage of our rail network and increase the useful life of more rail; however, our economic analysis of grinding indicates that this improvement is worthwhile only in certain corridors. When we invest in new rail in these high-density traffic corridors, we believe our investment includes future rail grinding costs. We face two choices, (1) buy rail only and get the purchase life, or (2) buy rail and improve it by grinding and realize a longer life. We achieve a longer life because grinding provides a benefit beyond the purchase life, and, therefore, grinding does not “merely reflect the return of the rail to its
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 14
original condition.” In addition to extending the life of the rail, grinding also improves safety and operating efficiency. As we discussed in our responses to prior comment letters, rail grinding improves safety by reducing derailments due to uneven rail wear, rail fatigue, and cracks and improves operating efficiency by enabling us to operate at higher train speeds, which increases our network capacity and reduces our need for locomotives, freight cars, and train crew employees.
With respect to your concern that rail grinding costs are not retired resulting in an overstatement of depreciable rail assets, we do retire grinding costs when rail is replaced. In this connection, we refer you to our response to comment number 4 of your letter dated August 31, 2009. There we included an example of the calculated retirement amount when one mile of rail in a high-density traffic corridor was replaced. In the example, “Cost of new rail” was the first variable in the calculation of average replacement cost. The “Cost of new rail” includes all amounts capitalized for rail in high-density corridors, including rail grinding. Therefore, amounts capitalized for rail grinding are included in historical costs and these costs are retired based on the amount of rail replaced.
We understand the SEC Staff’s references to accounting literature regarding the capitalization of rail grinding costs. We do not believe that it is appropriate to analogize capitalization of rail grinding to capitalization of costs to treat environmental contamination and, in particular, to removing rust from a chemical storage tank as discussed in Example 2.A of Exhibit 90-8A to EITF 90-8. Based on our review and assessment of EITF 90-8, if an analogy were to be made the more apt analogy is Example 2.B. Since we grind at acquisition, our activity is more akin to rust-coating than rust-cleaning. Thus, we believe that the literature you cite supports our accounting. We are further supported by FASB Concept Statement 6, as noted in previous correspondence.
Based on our conference call of January 5, 2010, we understand that the Staff has further considered its position stated in the above comment 7. We also understand the Staff’s view is that it is preferable to expense rail grinding costs as incurred; however, the Staff will not object to capitalization of rail grinding with an appropriate depreciation methodology. Upon reflection within the Company and in consultation with our independent registered public accounting firm, we have decided to voluntarily change our accounting policy for rail grinding beginning in the first quarter of 2010 from a capitalization method, under which we depreciate the capitalized costs, to a direct expense method, under which we will expense rail grinding costs as incurred. We acknowledge the fact that the direct expense method is preferable as it eliminates the subjectivity in determining the period of benefit associated with rail grinding. The Company’s independent registered public accounting firm has agreed with our proposal to account for this change as a change in accounting principle in accordance with SFAS 154. We will discuss this change in accounting principle in the footnotes to our 2009 Form 10-K. A preliminary draft of our proposed disclosure for inclusion in our 2009 Form 10-K is as follows:
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 15
Change in Accounting Principle for Rail Grinding Costs
The Company has historically accounted for rail grinding costs as a capital asset. Beginning in the first quarter of 2010, the Company intends to change its accounting policy for rail grinding costs from a capitalization method, under which the Company depreciated the capital costs, to a direct expense method, under which the Company will expense rail grinding costs as incurred. We believe the expense as incurred method is preferable as it eliminates the subjectivity in determining the period of benefit associated with rail grinding over which to depreciate the associated capitalized costs. The Company will reflect this change as a change in accounting principle from an accepted accounting principle to a preferable accounting principle in accordance with Accounting Standards Codification 250 – Accounting Changes and Error Corrections (formerly known as Statement of Financial Accounting Standard No. 154, Accounting for Changes and Error Corrections). The new accounting principle will be presented retrospectively to all periods presented in future earnings releases and SEC filings. When the accounting principle is retrospectively applied, net income for the years ended December 31, 2009, 2008, and 2007 will decrease by approximately $XX million, $XX million, and $XX million, or $.XX, $.XX and $.XX per share, respectively. The Company believes this change in accounting principle will not have a material impact on its consolidated financial position, results of operations, or cash flows.
You also suggest that we reevaluate all of our other capitalization policies regarding road assets. We evaluate our policies on a regular basis and believe that all of our capitalization policies are in accordance with U.S. GAAP. We regularly benchmark with the other major U.S. and Canadian railroads to ensure best and consistent practice. We went through a similar comprehensive review with the Staff last year on these and related issues. During the SEC comment letter process in 2008 and again in conjunction with the SEC comment letter process in 2009, we reassessed our accounting policies and evaluated compliance with these policies. Based on this assessment, we continue to believe our accounting policies are appropriate. However, our evaluation of compliance with these policies identified specific ballast cleaning costs (approximately $2 million during 2008) that had been capitalized although our policy is to expense these items as incurred. During our conference call of January 5, 2010, the Staff suggested that we consider additional disclosures regarding recurring maintenance-like costs that are capitalized. In light of the proposed change in our accounting for rail grinding costs and considering that we have not identified any additional recurring maintenance-like costs that are capitalized, we believe that we have addressed your concerns and that our proposed expanded disclosures referenced in comment 2 above are comprehensive.
| 8. | Refer to your response to our prior comment number 7. We note that track lining projects involve the addition of ballast to existing track structure to comply with Federal Railroad Administration track standards. We note further that you capitalize the costs attributable to track lining as you believe that the replacement of deteriorated, buried, and shifted ballast reflects both the installation of a new asset and the retirement of an old |
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 16
| asset. In this regard, please clarify for us whether the historical cost of ballast that is replaced as part of your track lining projects is removed from your gross property and equipment and accumulated depreciation balances in connection with new track lining projects. |
RESPONSE:
When a track lining project meets our minimum unit of property requirements and, therefore, qualifies for capitalization, the historical cost of the ballast replaced is retired from our gross property and accumulated depreciation balances.
* * * * *
As requested in your July 29, 2009 comment letter, we acknowledge that:
| • | The Company is responsible for the adequacy and accuracy of the disclosure in its filings; |
| • | Staff comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the filings; 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. |
Please feel free to call either me at (402) 544-6262 or Jim Theisen, Assistant General Counsel, at (402) 544-6765 if you should have any questions or further comments.
| Sincerely, |
| /s/ Jeffrey P. Totusek |
| Jeffrey P. Totusek Vice President and Controller Union Pacific Corporation |
| cc: | James R. Young |
| Chairman, President and Chief Executive Officer |
| Union Pacific Corporation |
| Robert M. Knight, Jr. |
| Executive Vice President-Finance and Chief Financial Officer |
| Union Pacific Corporation |
| Union Pacific Corporation Audit Committee |
Lyn Shenk, Branch Chief
Securities and Exchange Commission
January 11, 2010
Page 17
Appendix A
Roden, Andrew. “Slaves to the grind: rail grinding technology is developing rapidly to meet changing demands,” International Railway Journal, July 2007.
Roney, Michael. “Advances in rail grinding: experts in wheel and rail management are moving to a standard practice that calls for frequent rail grinding that reshapes rail to well-thought-out profiles,” Railway Track and Structures, June 2002.
Wu, Muimin and others. “Improvement of Effectiveness and Efficiency of Rail Grinding – A Rail Grinding Trial on a Norfolk Southern Railroad Line,” Technology Digest, September 2005.
Wu, Muimin. “Two key aspects in rail grinding-effectiveness and efficiency; researchers look at the effectiveness and efficiency of rail grinding to prolong rail life and reduce maintenance costs,” Railway Track and Structures, December 2004.
Zarembski, Allan. The Art and Science of Rail Grinding. Simmons-Boardman Publishing Company, 2005.