In-Process R&D in Business Acquisitions
More Disclosure Needed for Transparency and Comparability

By Nathan S. Slavin and Abu Ryan Khan

E-mail Story Print Story

FASB and the International Accounting Standards Board (IASB) are currently developing a joint project with a common exposure draft on accounting for business combinations. According to the current proposal, IPR&D will be measured at fair market value on the date of the business acquisition and capitalized as an intangible asset having an indefinite useful life. Upon completion of the research project, it must be either expensed as a failed project or capitalized as an amortized intangible asset. This proposal is a drastic departure from the prevailing accounting rules, which require the immediate expensing of all recognized IPR&D assets following a business combination.

The authors studied a random sample of 380 business combinations from 1996 to 2004 where an IPR&D charge was reported. From the study, they found that since the SEC in 1998 expressed its intention to scrutinize this area—

• companies are taking lower IPR&D charges;
• companies are recognizing more goodwill; and
• companies are recognizing the same proportion of IPR&D and goodwill since the issuance of SFASs 141 and 142.

Due to the complicated nature of IPR&D and the lack of disclosure requirements, one cannot be certain that it is not still used for earnings management. The SEC’s scrutiny, which began in 1998, and previous studies have focused on the magnitude of the IPR&D charge. The magnitude of an IPR&D charge does not, however, always provide useful information about its earnings potential. This study demonstrated that certain underlying assumptions made by management can significantly affect the valuation of an IPR&D project. Specifically, the pattern of the projected cash flows generated from IPR&D projects and the discount rate selected have an enormous effect on the resource’s value. The authors suggest that today’s disclosure requirements on IPR&D charges, which do not require companies to disclose important management assumptions, are inadequate and should be addressed by FASB in its exposure draft on business combinations.

Current Accounting Rules for IPR&D

Purchased IPR&D represents the estimated fair value assigned to R&D projects, acquired in a business combination (purchase method), that have not been completed at the date of acquisition and that have no future alternative use. SFAS 2, Accounting for Research and Development Costs, issued in 1975, required companies to expense R&D costs in the period in which the costs were incurred. This created a special problem for business combinations. An acquiring company often purchases complete and incomplete research and development projects with substantial market value. FASB subsequently clarified the accounting for the acquired completed and in-process research projects by issuing FASB Interpretation 4 (FIN 4), Applicability of FASB Statement No. 2 to Business Combinations Accounted for by the Purchase Method—an Interpretation of FASB Statement No. 2, in 1975.

FIN 4 recognizes two categories of R&D-related assets in a business combination:

all tangible and intangible assets resulting from research and development activity of the acquired enterprise …. Identifiable assets resulting from the research and development activity of the acquired enterprise might include, for example, patents received or applied for, blueprints, formulas and specifications or designs or new products or process.

[All tangible and intangible assets] … to be used in research and development activity of the combined enterprise …. Identifiable assets to be used in research and development activity of the combined enterprise might include materials and supplies, equipment and facilities and, perhaps, even a specific research project in-process. [italics in original]

In the purchase method of accounting for business combinations, both categories of assets associated with R&D activities should be assigned costs determined from the price paid by the acquiring entity and not from the historical cost of the acquired entity (as restated in SFAS 141). However, the subsequent reporting for these two categories of assets is different. The first category of tangible and intangible assets resulting from R&D activity must be valued at fair market value and capitalized as an amortized intangible asset. The second category of recognized assets (to be used in R&D) must be expensed in the period of the business combination as “in-process research and development” expense.

The authors obtained an example of an IPR&D expense from the 10-Q of Cephalon Corp., a pharmaceutical company, for the quarter ending September 30, 2004. On August 12, 2004, Cephalon Corp. acquired a 100% interest in CIMA Laboratories (another pharmaceutical company) for a total cost of $514.13 million, and recognized an IPR&D expense of $185.7 million. The following notes from the quarterly report disclosed the IPR&D expense:

We allocated $185.7 million of the purchase price to in-process research and development projects. In-process research and development (IPR&D) represents the valuation of acquired, to-be-completed research projects ... The estimated revenues for the in-process projects are expected to be recognized from 2006 through 2019 … The value assigned to purchase in-process technology was determined by estimating the costs to develop the acquired technology into commercially viable products, estimating the resulting net cash flows from the projects, and discounting the net cash flows to their present value. The revenue projections used to value the in-process research and development were, in some cases, reduced based on the probability of developing a new drug, and considered the relevant market sizes and growth factors, expected trends in technology, and the nature and expected timing of new product introductions by us and our competitors … [D]iscount rates of 28 percent were considered appropriate for the in-process research and development. These discount rates were commensurate with the projects’ stage of development and the uncertainties in the economic estimates described above …. At the date of acquisition, the development of these projects had not yet reached technological feasibility, and the research and development in progress had no alternative future uses. Accordingly, these costs were charged to expense in the third quarter of 2004.

Cephalon also provided in the 10-Q a computation and allocation of the difference between the purchase price of the acquired company and the fair value of the acquired net assets (Exhibit 1).

IPR&D’s Value for Earnings Management

In a 1996 Barron’s article, the renowned accounting educator Abraham Briloff reported how an IPR&D charge of $1.8 billion dollars related to IBM Corp.’s acquisition of Lotus Corp. affected IBM’s earnings in subsequent years. Briloff characterized IBM’s IPR&D charge as analogous to purchasing inventory and immediately expensing it in the period without a concomitant sales transaction. The income reported in subsequent periods would be significantly overstated because the costs related to the inventory would not be matched when it would be sold. The authors’ position is that an IPR&D project satisfies the definition of an asset under FASB Concept Statement 6, Probable Future Economic Benefits Obtained or Controlled by a Particular Entity as a Result of Past Transactions or Events. An IPR&D project is purchased with the expectation that it will generate income in the future. The value of an IPR&D project includes the present value of probable incremental cash flows from the project. Under current GAAP, the value assigned to the IPR&D project is expensed on the date of the business acquisition. This relieves all future revenue generated from the project from any costs assigned and expensed at the acquisition date.

The probable incremental cash flows from IPR&D assets are often material to the acquiring company’s annual income. To illustrate the potential income that could be generated from an IPR&D asset, the authors examined the IPR&D charge taken by Cephalon above. Ignoring amortization, the authors calculated that the IPR&D that was written off in 2004 could potentially contribute $53,690,162 annually to Cephalon’s after-tax income for 14 years. The authors assumed the following to calculate potential annual after-tax income from the written-off IPR&D asset:

• Present value of IPR&D = $185,700,000
• Discount rate = 28%
• Number of periods = 14
• Amortization amount = $0

The potential $53.7 million net income from IPR&D represents 64.02% of the company’s 2003 net income ($83.86 million), and 52.44% of its 2004 pro forma net income. Cephalon’s 2004 pro forma income of $102.38 million was computed by adding back the IPR&D charge of $185.7 million to its income before taxes, assuming a tax rate of 35%. If Cephalon’s revenue and expense projections related to its acquired IPR&D projects materialize, the company’s performance ratios (e.g., return on equity and operating margin) will dramatically improve in future reporting periods. This potential improvement will most likely be attributed to Cephalon’s management, although it would have been caused by accounting methods and assumptions used when the IPR&D project was purchased.

An IPR&D charge can be a powerful management tool for generating future earnings improvement. In a 1998 study, Zhen Deng and Baruch Lev observed that companies improved their earnings on an average by 25% and their return on equity by 37% when IPR&D projects are expensed rather than capitalized as amortized assets. An IPR&D charge may signal future earnings growth to investors. Deng and Lev found that investors value the acquired R&D differently from goodwill. They concluded that investors perceive, on average, the fair market values assigned to IPR&D projects as credible indicators of asset values. A company with a larger IPR&D charge is more likely to be rewarded by investors in postacquisition periods.

IPR&D in the Past Deng and Lev found only three

IPR&D charges that occurred during the 1980s, and 372 IPR&D charges from 1990 to 1996. Of those IPR&D charges, 147 occurred in the first half of 1996. They found that investors had greater confidence in a business acquisition when a larger proportion of the purchase price was allocated to IPR&D and a smaller proportion was allocated to goodwill. The mean IPR&D charge was $29.76 million, representing 72% of the purchase price. In the first three quarters of 1998, U.S. companies expensed $14.8 billion of IPR&D projects recognized in business acquisitions. In 1999, then–SEC chief accountant Lynn Turner described the magnitude of these IPR&D charges as “just unreal.”

In late 1998, the SEC began addressing the area of IPR&D charges as well as corporate restructuring charges. The SEC examined nearly 50 companies and $10 billion in IPR&D charges in the fourth quarter of 1998. As a result of this examination, nearly $5 billion of the write-downs were reversed. In addition, three large accounting firms—Pricewaterhouse Coopers, Deloitte & Touche, and KPMG—agreed to discontinue assisting corporate audit clients involved in IPR&D projects and other one-time write-offs.

In their 2004 study, Dowdell and Press measured the mean IPR&D charge as a proportion of the purchase price for 582 acquisitions in the computer-programming and software industry in the period from 1996 to 2001. The mean IPR&D charge after the SEC’s intervention in the fourth quarter of 1998 declined from 63% to 18%, a statistically significant decrease of 71%. The study also examined 71 companies that restated prior financial statements related to 93 business acquisitions. The amount charged to IPR&D was restated from 66% to 25% of the purchase price, a statistically significant decline of 62%. The results of the study suggest that the SEC’s intervention had a significant impact on the amount allocated to IPR&D in a business acquisition.

The Current Study

The authors’ objective was to expand and generalize the previous studies by examining a more diverse and recent random sample of business acquisition cases with IPR&D charges. Dowdell and Press’ sample was limited to the computer-programming and software industry. In addition, the release of SFASs 141 and 142 in 2001 meant that goodwill is no longer amortized. Companies might have an incentive to allocate a larger proportion of the purchase price of a business acquisition to this asset and a smaller proportion to IPR&D. The authors examined whether this new incentive affected the valuation of IPR&D.

To analyze the effects of the SEC’s scrutiny and SFASs 141 and 142 on IPR&D, the authors identified the following periods:

• Period 1: January 1, 1996, to September 9, 1998 (the day Lynn Turner sent a letter on IPR&D charges to Robert Herz, then-chairperson of AICPA’s SEC Regulations Committee)
• Period 2: September 10, 1998, to June 30, 2001 (the month SFASs 141 and 142 were issued)
• Period 3: July 01, 2001, to December 31, 2004.

The authors anticipated that the allocation of acquisition cost to IPR&D and goodwill in their sample of companies from different industries would be similar to the results found in the previous studies. In addition, they anticipated finding an inverse relationship between the amount allocated to IPR&D and goodwill, under the assumption that the valuation assigned to all other tangible and intangible assets would not be modified.

Sample Selection and Characteristics

The business acquisition cases examined were completed between January 1, 1996, and December 31, 2004. An IPR&D charge had to be taken during the business acquisition, and adequate details of the IPR&D charge had to be available from the acquiring company’s 10-K filing.

The authors used the Edgar Online Pro database to search 10-K filings of companies, using carefully chosen keywords. They randomly selected 100 business acquisitions from each year and reviewed the financial statements and the related footnotes for any IPR&D charge. After excluding 520 filings that did not provide adequate information about the IPR&D charge and acquisition cost, the authors arrived at a sample of 380 business acquisitions that reported charges for IPR&D. They collected the following information from the 10-K filings of this sample:

• Acquisition cost;
• Amount allocated to IPR&D;
• Amount allocated to goodwill; and
• Amount allocated to other intangibles.

From this data, the authors calculated the proportion of the acquisition cost allocated to IPR&D. They also determined the proportion of the acquisition cost allocated to goodwill.

Findings

The findings of the study were as follows:

• The mean proportion of acquisition cost allocated to IPR&D decreased significantly after the SEC’s scrutiny in late 1998.
• The mean proportion of acquisition cost allocated to IPR&D did not change significantly after the issuance of SFASs 141 and 142.
• The mean proportion of acquisition cost allocated to goodwill increased significantly after the SEC’s scrutiny in late 1998.
• The mean proportion of acquisition cost allocated to goodwill did not change significantly after the issuance of SFASs 141 and 142.

The results of the study are summarized in Exhibit 2. The mean proportion of acquisition cost allocated to IPR&D declined by more than half, from 56.45% in Period 1 to 24.94% in Period 2. This is consistent with the results found by Dowdell and Press. As expected, because of the decrease in the mean proportion reported for IPR&D, the mean proportion of acquisition cost allocated to goodwill increased dramatically, from 12.81% in Period 1 to 35.44% in Period 2.

The results of the study suggest that the SEC’s intervention significantly reduced the valuation of IPR&D in business acquisitions. The mean proportion of acquisition cost allocated to IPR&D decreased slightly from 24.94% in Period 2 to 24.80% in Period 3. The amount allocated to goodwill, however, continued to increase from Period 2 to Period 3, rising from 35.44% to 42.79%. This suggests that the introduction of SFASs 141 and 142 was a nonevent for the reporting and measurement of IPR&D. The effect of SFASs 141 and 142 on goodwill is beyond the scope of this study. The sample chosen was not representative of all business combinations, but of business combinations that involved IPR&D charges.

Valuation of IPR&D: Subjective and Discretionary

The initial focus of this study was the effect of the SEC’s scrutiny and the influence of SFASs 141 and 142 on IPR&D valuations. As a by-product of the findings, the authors learned that management has considerable discretion in estimating the variables and components used to value IPR&D projects. The authors anticipated that the amount allocated to IPR&D would be the only factor affecting a company’s future income; that is, a lower IPR&D charge would result in less future income. The study found that the income potential of an IPR&D asset cannot be predicted only from the nominal value assigned to this resource. The magnitude of an IPR&D charge is not the only measure of its hidden future earnings potential; other key assumptions, such as discount rates and estimated future cash flows, have a significant effect. It is possible to achieve similar future earnings under a range of values assigned to IPR&D.

The Valuation Process of IPR&D Projects

IPR&D valuation is generally based on the following guidelines:

• Identification of assets. FIN 4 states that all tangible and intangible assets “to be used in research and development activity of the combined enterprise” must be expensed as acquired IPR&D. The intent of the combined enterprise is, therefore, crucial to the identification of acquired assets to be used in research and development projects. Consequently, the fair market value of the acquired facility can be expensed in the period of acquisition.
• Valuation of assets. FIN 4 requires acquired R&D to be valued according to APB Opinion 16 (1970). Under APB Opinion 16, as restated by SFASs 141 and 142, an acquiring firm must allocate the acquisition price to identifiable tangible and intangible assets. In practice, most IPR&D valuations are estimated at the present value of the expected incremental cash flows from this purchased resource.

The expected incremental cash flows from an acquired IPR&D project and the appropriate discount rate for the project are estimated considering a variety of factors. A presentation by the American Society of Appraisers recommended a multifactor analysis of data to forecast future economic benefits of an IPR&D project. Appraisers were advised to consider factors such as market demand, market share, competitive products and technology, expected market penetration, current stage of the project, estimated amount of effort already expended on the project, and the expected life of the technology or product. A sales forecast is normally prepared incorporating all of these. The sales forecast is then reduced by the cost of goods sold, sales and administrative expense, and taxes, to arrive at net income. Amortization and tax benefits are added to the net income, resulting in an incremental cash flow for each period. The IPR&D value is then estimated by calculating the present value of expected incremental cash flows from the project, using a discount rate reflecting the risk of the research project. These complexities make the valuation of IPR&D inherently difficult and subjective.

Further Disclosure Needed

Companies today are not required to disclose the expected cash flows from IPR&D assets, or the discount rate used to calculate the value of IPR&D. Few companies voluntarily disclose key valuation assumptions. Only 55 of the 380 business acquisitions included in this study (14.5%) disclosed the discount rate used. The rates ranged from 13% to 60%, with an mean rate of 28.8% and a median rate of 30%. Fewer companies provided any information on their cash flow assumptions.

Cephalon was one of the few companies that disclosed some key assumptions used to value the IPR&D projects recognized in an acquisition. In its acquisition of CIMA Lab, Cephalon projected 14 years of annual cash flows (2006 to 2019) and used a discount rate of 28% in calculating the $185.7 million charged to IPR&D.

To demonstrate the effect of cash flow assumptions on the value of the recognized IPR&D asset (Exhibit 3), the authors considered three cash flow streams, assuming the same total cash flow and a discount rate of 28%. In each scenario, both the amount of the IPR&D charge and its proportion of acquisition cost change drastically. The value of the IPR&D charge could range from $46.43 million (9.03% of the acquisition cost) to $371.40 million (72.24% of the acquisition cost), depending upon the cash flow scenario assumed.

To evaluate the effect of discount rate assumptions on the value of the recognized IPR&D asset, the authors considered three discount rate alternatives, assuming an even cash flow of $53.69 million per year for 14 years (Exhibit 4). The authors used alternative discount rates of 14%, 28%, and 40% to demonstrate the range of values that could be assigned to IPR&D. The value of the IPR&D charge could range from $133.02 million (25.87% of the acquisition cost) to $322.25 million (62.68% of the acquisition cost), depending upon the discount rate assumed.

When the projected pattern of the cash flow streams and the discount rate are used as a set, the range of the amounts assigned to the IPR&D project is even more pronounced. A conservative set of assumptions—increasing annual cash flow (Exhibit 3, Scenario A) and a large discount rate of 40% (Exhibit 4, Scenario C)—would result in an IPR&D charge of $33.25 million, or 6.47% of the acquisition cost. Conversely, the least conservative set of assumptions—decreasing annual cash flow (Exhibit 3, Scenario C) and a small discount rate of 14% (Exhibit 4, Scenario A)—would result in an IPR&D charge of $644.5 million, or 125.36% of the acquisition cost.

These alternative scenarios demonstrate the powerful effect that assumptions have on the value assigned to IPR&D projects. The authors’ position is that merely disclosing the magnitude of the IPR&D charge will not inform the investing public of the income-generating power of the IPR&D resource.

Impact of the New Proposed Standard

On June 30, 2005, FASB issued exposure draft 1204-001 to modify the accounting standards for business combinations. Paragraph D-22 of the exposure draft will require capitalizing the value allocated to IPR&D as an asset and amortizing it over its useful life once the project is completed. The exposure draft amends paragraph 16 of SFAS 142 as follows:

Intangible assets that are acquired in a business combination for use in a particular research and development project and that have no alternative future uses shall be considered indefinite-lived until the completion or abandonment of the associated research and development efforts, at which point the acquirer would make a separate determination of the useful life of that asset.

This would significantly alter the accounting of IPR&D. Under the proposal, IPR&D assets will be initially measured and capitalized at fair value as an intangible with an indefinite life during business combinations. In every reporting period, management will make a determination of whether the project can be classified as “in process.”

Once management determines that the project is no longer in process, because it has either failed or succeeded, it must reasonably estimate the useful life of the IPR&D asset. In the event of a failed project, the IPR&D asset will be considered impaired, and expensed in the period of impairment. In the case of a successful IPR&D project, the asset will be amortized over its useful life. As a result, revenue will be matched to the amortization expenses from the IPR&D asset. Under the proposed rule, the reporting of IPR&D projects is similar to that of other intangible assets.

The proposed rule provides several benefits over the current rule for IPR&D:

• Companies will not be able to use IPR&D as an immediate charge to income in the period of a business acquisition.
• The proposed rule will require management to write off an IPR&D project only if a decision has been made to discontinue it.
• The proposed rule will require that the income generated from the IPR&D projects be matched to scheduled amortization expenses.
  

The authors suggest that, under the proposed rule, management will have an incentive to recognize a lesser amount of IPR&D, for two reasons:

• If the project fails, the impairment write-off will be smaller.
• If the project succeeds, the amortization expense from the IPR&D asset will be smaller, making the income generated greater.

Notwithstanding the new provisions, the exposure draft does not address the need for disclosing the assumptions used in the valuation of IPR&D projects. The authors believe that additional disclosure regarding IPR&D valuation will enhance the transparency of this issue.

The authors have previously demonstrated that a very conservative set of assumptions, consisting of an increasing cash-flow pattern and a large discount rate, will result in a diminished value assigned to the IPR&D project. Without minimal disclosure of discount rates and cash flow assumptions, the value allocated to IPR&D may not be comparable across similar business acquisitions. IPR&D projects with the same nominal value may result in very different future income streams.

The example illustrated in Exhibit 5 demonstrates this point. Assume that similar companies, P1 and P2, purchase companies S1 and S2, respectively. P1 recognizes an IPR&D asset of $200 million, using a discount rate of 14% and even cash flows from 2007 to 2011. P2 recognizes an IPR&D asset of $200 million, using a discount rate of 28% and even cash flows from 2007 to 2011. In the beginning of 2007, both IPR&D projects are successful, managements’ cash flow assumptions are proved correct, and both assets’ useful lives are estimated at five years.

Although both IPR&D assets have the same nominal value and the same amortization expense of $40 million per year, Company P1’s IPR&D asset will generate an annual income of $18.26 million, while Company P2’s IPR&D asset will generate an annual income of $38.99 million. The difference in income is attributable to the discount rate chosen.

The current exposure draft reduces disclosure requirements in connection with IPR&D. Paragraph B186(f) of the exposure draft states:

Eliminated the requirement to disclose the amount of in-process research and development acquired and had been measured and immediately written off to expense in accordance with Statement 141 [paragraph 51(g)]. This Statement no longer permits that past practice.

FASB considered this disclosure unnecessary because IPR&D will no longer be expensed. IPR&D will be initially classified as a nonamortized intangible asset. The authors believe the proposed rule does not recognize the unique nature of an IPR&D asset. The draft could enhance transparency by requiring management to disclose the cash flow and discount rate assumptions used in the valuation of the IPR&D projects. In addition, this would enable investors to measure the potential income from a successful IPR&D asset. It would also make financial statements more comparable.

Disclosure Still Needed

The initial focus of this study was to determine how two events—the SEC’s increased scrutiny of IPR&D charges, and the new accounting rules under SFASs 141 and 142—affected the valuation of IPR&D. The authors found that the SEC’s intervention significantly diminished the proportion of acquisition cost allocated to IPR&D, while SFASs 141 and 142 had no significant influence.

As a by-product of this study, the authors discovered that the assumptions used to value IPR&D projects are inadequately disclosed in companies’ annual reports following a business acquisition. A new FASB exposure draft reduces the disclosure requirements even further, although the authors believe that more disclosure is necessary for transparency and comparability. The authors’ position is that although these IPR&D projects will be capitalized and reported as amortized assets after they are completed, the disclosure of the assumptions used to value theses resources will still be needed by investors.