Accounting for Asset Retirement Obligations
Understanding the Financial Statement Impact

By Robert E. Guinn, Richard G. Schroeder, and Suzanne K. Sevin

E-mail Story Print Story

When SFAS 143 was issued, concern was voiced over the economic consequences to the companies that its provisions affect. Some suggested that companies would be required to record materially increased expenses as well as increases in assets and liabilities that might be large enough to negatively impact commonly used financial ratios such as the return on assets, debt-equity ratio, and operating profit ratio. This study examines the financial statement effect of the adoption of SFAS 143 on a sample of public companies.

SFAS 143’s Objective

The objective of SFAS 143 is to establish accounting standards for the recognition and measurement of obligations attributable to the removal of assets as well as to their associated restoration costs. This obligation is to be recorded in the period in which the liability meets FASB’s definition of a “probable future sacrifice of economic benefits arising from a present obligation,” and in which its amount can be reasonably measured. Uncertainty with respect to the timing or method of settlement that is conditional on future occurrences does not affect the recognition of the liability but may be factored into its measurement.

The obligation must be recorded at fair value, which represents the amount that a third party would require to assume the liability. If an active market for these obligations does not exist, the company must use the expected present value technique outlined in Statement of Financial Accounting Concepts 7, Using Cash Flow Information and Present Value in Accounting, which results in measuring the asset’s and related liability’s present value by using each company’s credit-adjusted rate.

When a company first adopts the standard, a liability is recorded for the present value of the future removal of the applicable asset as well as for the associated restoration costs. The asset’s carrying value is increased by the capitalized cost and, depending on prior accounting treatment, further adjusted for previously recorded depreciation. The offsetting adoption entry is recorded as a change in accounting principle (CAP) cumulative effect adjustment and is reflected on the income statement.

Asset and liability amounts under SFAS 143 result in depreciation and accretion expenses each year. The capitalized asset retirement cost is allocated in a systematic and rational manner as depreciation expense over the estimated useful life of the asset. The depreciation charge is calculated as the increased asset base divided by the asset’s useful life on the date of adopting SFAS 143. Accretion expense is calculated by multiplying the balance of the recorded liability by the company’s credit-adjusted discount rate each year, and is simply the amortization of the present value discount associated with the ARO’s initial recording. Accretion expense is offset with an increase to the liability account, and, at the end of the asset’s life, the liability account will have a balance equal to the amount needed to settle the retirement obligation. If any of the original assumptions change, a recalculation of the obligation and subsequently associated expenses should be recorded as a change in accounting estimate, which is then offset by an increase in the asset’s value, less any accumulated depreciation.

Disclosure requirements of SFAS 143 include the following:

• A description and the amount of increase of the ARO liability and associated asset;
• The amount of the change in accounting principle resulting from the adoption of the new standard; and
• A reconciliation of the beginning and ending carrying amounts, showing separately other changes attributable to:
• Liabilities incurred in the current period;
• Liabilities settled in the current period;
• Accretion expense; and
• Revisions in estimated cash flow whenever there is a change in one or more of the components during an accounting period.

Accounting for SFAS 143 Adoption

To illustrate the accounting for application of SFAS 143, assume a company purchased an asset with a useful life of 25 years for $500,000 on January 1, 1985, and assume that the company had not established a provision for retirement costs in valuing its assets and liabilities prior to the effective date of SFAS 143. On January 1, 2003, when the standard became effective, the company must use current information to assess whether an ARO exists. Exhibit 1 indicates that an ARO of $175,000 existed on the date the asset was acquired. According to SFAS 143, the present value of this amount should have been capitalized and recorded as a liability on January 1, 1985. Using the company’s credit-adjusted discounted rate of 13% would have resulted in an ARO of $8,243 being recorded on January 1, 1985, when the asset was acquired, with a corresponding increase in the capitalized retirement cost of $8,243. Subsequently, the company should have recorded annual accretion expense of 13% of the current carrying value of the liability and additional depreciation expense of $330 each year.

Exhibit 1 illustrates the result of these adjustments, assuming no accounting treatment for asset retirement obligations prior to adopting SFAS 143. Upon adopting the standard on January 1, 2003, an asset book value of $142,303 (which is $2,303 greater than the amount of the book value of the asset without the ARO cost) and a related liability of $74,386 are recognized. Recording these amounts on the effective date of SFAS 143 results in an increase in the book value of assets of $2,303, which consists of the difference between capitalized retirement costs of $8,243 and an increase in accumulated depreciation of $5,940, an increase in ARO of $74,386, and a negative cumulative effect of a change in accounting principle of $72,083. Thereafter, depreciation expense is increased by $330 for each of the remaining seven years of the asset’s life, and accretion expense is recorded each year by multiplying the carrying value of the ARO by the company’s credit adjusted interest rate of 13%. Exhibit 1 discloses the result of these calculations and the journal entry for the adoption of SFAS 143 on January 1, 2003.

The cumulative effect of the change in accounting principle is dependent on whether companies have recognized and depreciated estimated retirement costs prior to adopting SFAS 143. As indicated in Exhibit 1, the company is assumed not to have recorded or depreciated retirement costs from 1985 through 2002, and thus, the cumulative effect of adopting of adopting SFAS 143 on January 1, 2003, resulted in a loss of $72,083. This loss reflects the catch-up effect related to depreciation and accretion expenses for the prior 18 years. Exhibit 2 shows that if the company had capitalized and depreciated retirement costs of $175,000 during the first 18 years, and recorded the liability at its total future amount, the cumulative effect of adopting SFAS 143 would be a gain of $53,917. This gain results from the differences between depreciation on the undiscounted retirement costs ($175,000) and the depreciation and accretion expenses relative to the discounted retirement cost and the related liability ($8,243).

Survey Objectives

SFAS 143 became effective for all companies whose fiscal years began after June 15, 2002, and consequently, December 31, 2003, was the first reporting date for most companies under its provisions. To determine the impact of SFAS 143, the authors selected a random sample of 1,000 U.S. public companies from the Russell 300 Index (excluding non–calendar year-end companies and financial institutions). The research addressed the following questions:

• What percentage of U.S. companies was impacted by SFAS 143 adoption at December 31, 2003?
• What industries were most affected by SFAS 143 adoption?
• What is the financial statement impact of required SFAS 143 accounting as a—
  • percent of total assets?
  • percent of total liabilities?
• What is the impact on earnings of recording accretion expense?
• What is the impact on earnings of recording the cumulative prior effect associated with SFAS 143?

Results

The 2003 SEC Form 10-K reports for each of the 1,000 sample companies were reviewed to determine if they included AROs. Exhibit 3 indicates that 105 sample companies (10.5%) were affected by the adoption of SFAS 143 for their existing asset retirement obligations and that the affected companies were concentrated in only a few industries [classified by Standard Industrial Classification (SIC)]. The majority were in metal and mining (58) and in transportation and
utilities (36).

Exhibit 4 provides an indication of the impact of SFAS 143 on the surveyed companies’ financial statements. Reviews of financial statements revealed that some companies did not provide all of the information required by SFAS 143. Ninety-seven of the 105 companies disclosed the amount of their AROs, and 83 disclosed the increase in the ARO-related asset. Eighty-two of the 97 sample companies that reported ARO amounts disclosed the amount of 2003 accretion expense, and 86 disclosed the amount of the change in accounting principle resulting from SFAS 143 adoption. It is possible that those amounts that were not disclosed may have been considered immaterial by management.

As reported in Exhibit 4, Panel A, of the 97 sample companies that reported changes in liabilities associated with SFAS 143 adoption, 93 reported an increase, while four reported a decrease. Of the 83 sample companies that reported changes in assets, 81 reported an increase, and two reported a decrease. Additional data suggest that, during 2003, the average increase in liabilities associated with reporting AROs was 4.0%, with a range of 0.1% to 34.8%, and that the average increase in ARO-related assets was 2.6%, with a range of 0.1% to 29.8%.

Exhibit 4, Panel B, reports the impact of SFAS 143 adoption on the 2003 income statements of sample companies, specifically with respect to accretion expense as a percentage of income from continuing operations and with respect to the associated CAP as a percentage of net income. Eighty-two companies disclosed the information necessary to determine SFAS 143’s impact of accretion expense on 2003 income. Of these, 71 reported positive income from continuing operations, while 11 reported negative income from continuing operations. For companies with positive income from continuing operations, the average impact of 2003 accretion expense was 6.7%, with a range of 0.1% to 86.7%. For those companies reporting a net loss from continuing operations, the impact of accretion averaged 5.1%, with a range of 0.1% to 19.5%.

Exhibit 4 also reflects that 86 of the 105 sample companies reported a cumulative effect from a CAP as the result of adopting SFAS 143. Forty-four of the 86 reported a positive impact on earnings, with an average impact of 8.9% and a range of 0.1% to 65.7%. Forty-two of the 86 reported a negative impact on earnings. The average impact for those companies was 16.7%, with a range of 0.1% to 96.7%.
The average increases in assets and liabilities were 2.6% and 4.0%, respectively. Further untabulated analysis of the data indicates that these results are driven by a few companies that had very large increases in assets and liabilities, as evidenced by (unreported) median increases in assets and liabilities of 1.3% and 2.6%, respectively.

While the percentage increases in expenses are somewhat higher, further examination of the results leads to a similar conclusion. The average accretion impact of approximately 6% was distorted by the results reported by a few companies. The median reported accretion expense as a percentage of income from continuing operations was 2.1%.

SFAS 143 does not require companies to specifically identify the change in depreciation expense attributable to the standard’s adoption. None of the companies surveyed disclosed this information, and therefore the impact of depreciation expense on income relative to capitalized retirement costs could not be specifically determined. It is possible, however, to make some observations concerning the depreciation effect.

The relationship of each company’s accretion and depreciation expenses will depend on the age and life of its assets, as well as on the company’s credit-adjusted rate. When attempting to assess the change in depreciation charges, the situation illustrated in Exhibits 1 and 2 may not be representative of the companies contained in the sample, for two reasons. First, the depreciation period is probably much longer than 25 years. (For example, nuclear power generators are generally depreciated over 40-year periods.) Second, the interest rate used in the example, 13%, may be lower or higher than the actual credit-adjusted rates applicable to the sample companies. When valuing its AROs upon the adoption of SFAS 143, a company must use its credit-adjusted discount rate on the date the asset was first acquired, and accordingly, the rate applied is influenced by the age of the applicable assets. Interest rates have declined substantially over the past 25 years. The 30-year Treasury bond rate has decreased from 11.98% at December 31, 1980, to 5.46% at December 31, 2000.

The average accretion expense reported by the sample was 12.8%, suggesting that the assets subject to SFAS 143 reporting were acquired in the early 1980s, and providing the rationale for the scenario set forth in Exhibits 1 and 2. Exhibit 5, Panel A, reflects the incremental amount of depreciation and accretion expenses attributable to SFAS 143 adoption, based on the Exhibit 1 data.

This exhibit indicates that when capitalized retirement costs are allocated over long periods of time and the depreciation rate is lower than the discount rate, the incremental depreciation expense associated with the standard will be less than the applicable accretion expense, and that the difference widens over time. Because the credit-adjusted discount rate used in the calculation has an effect on incremental depreciation and accretion expenses, Exhibit 5, Panel B, reflects the same scenario, assuming a credit-adjusted rate of 8%. Lower discount rates result in a larger amount of capitalized retirement cost and, therefore, a higher amount of annual depreciation. Additionally, the initial ARO will be relatively higher, resulting in lower total accretion expense over the ARO’s estimated life. Due to these combined effects, a lower discount rate will also cause the cumulative effect of a change to SFAS 143 to result in a larger loss or a smaller gain.

Discussion

The results of this study indicate that much of the concern over the SFAS 143 requirement to capitalize AROs and to record additional expenses appears to be unfounded. Findings based on a random sample suggest that only approximately 10% of all U.S. public companies are affected by the standard’s reporting requirements, and that, on average, the companies affected by this pronouncement reported relatively minor changes in assets, liabilities, and expenses. Some companies, however, did experience significant financial reporting effects, both positive and negative. For example, one coal mining company recorded an 86% positive change on income resulting from the change to SFAS 143, and a $44 million (27%) decrease in its recorded liability due to changing the measurement of the expected liability from an undiscounted amount to its present value. For other companies, the impact was significantly negative. One oil and gas extraction company recorded a 10% increase in its liabilities, a 6% increase in its assets, and a 15% increase in its expenses due to SFAS 143, because the company had previously recorded the estimated costs of AROs as part of depreciation and depletion without recording a separate liability for the amounts. The companies most significantly affected by SFAS 143 appear concentrated in the coal mining, oil and gas mining, and oil and gas services industries.

A particularly interesting finding is the fairly even percentage of companies experiencing increasing versus decreasing cumulative effect adjustments. As Exhibit 1 illustrates, income-decreasing CAPs result when no provision had previously been made in the accounting records for those retirement costs, while income-increasing CAPs result primarily when companies had previously included the asset’s retirement costs in the depreciable base (Exhibit 2). Forty-seven percent (42 companies) reported a positive impact on earnings associated with the standard, suggesting that many companies had incorporated retirement obligations into their accounting records prior to the issuance of SFAS 143.

A close examination of the requirements of SFAS 143 and the results of the study indicate that the proper disclosure of AROs is a complex issue. Inherent in the calculation of the ARO and its related asset cost are numerous assumptions and judgments, including the estimated life of the property to be retired, settlement amounts, inflation factors, credit-adjusted discount rates, timing of settlement, and changes in the legal, regulatory, and environmental landscapes. These assumptions and judgments must be carefully evaluated to ascertain their appropriateness.