Corporate taxpayers undergoing certain tax-free transactions now have several more options to determine their tax basis in the acquired Target stock. Rev. Proc. 2011-35, which was issued by the Internal Revenue Service on May 31, 2011, and became effective June 20, 2011, updates 30 years of basis determination procedures for basis-transfer transactions.
If a company has engaged in a tax-free stock acquisition or if a transaction is in the planning stages, a tax executive should consider the shifting landscape of target tax basis determination guidelines. This article reviews the history of basis determination techniques and the evolution of guidance from Rev. Proc. 81-70 to Rev. Proc. 2011-35. It also explores the potential hazards of delaying basis determination and provides best practices for identifying the most appropriate method for determining target stock tax basis.
Stock, a common vehicle for corporate acquisitions, has been the sole or part consideration in one third of all U.S. transactions since 1996.1 When at least 80 percent of a company’s stock is acquired by another company in an exchange solely for the Acquiring Company’s voting stock, the transaction may qualify as a type B reorganization (under section 368(a)(1)(B) of the Internal Revenue Code) and qualify as tax-free.
In a B reorganization, the acquirer’s tax basis in the newly acquired Target stock is the surrendering transferor’s basis in the stock (or carryover basis under section 362(b)). The basis is carried over from transferor to transferee; such is the case with a basis transfer transaction.
Another basis transfer transaction is the Section 351 merger, where typically an acquirer and target merge to form a “NewCo” company. To be considered tax-free, at least 80 percent ownership by vote and value of each of the target and acquirer’s stock must be controlled by the NewCo after the merger.
An example of the determination of target stock tax basis in a basis transfer transaction is as follows: Assume Company T has five shareholders with 10 shares each and a tax basis of $10 per share, for an aggregate $500 in basis (5 x 10 x $10). In a B reorganization, Company A uses voting stock to acquire 100 percent or all 50 shares of Company T. Company A’s basis is transferred from the surrendering holder, such that A’s tax basis in the acquired T stock will be $500, or Company T shareholders’ stock aggregate basis.
If target stock is publicly traded, ascertaining carryover basis can be quite complicated since a significant portion of the transferor shareholders may not be easily identified. The complexity behind target basis determination in basis transfer transactions is typically a function of these issues: (1) the number of issued and outstanding Target Company shares, (2) the number of transferor shareholders, and most critically (3) the degree the transferred shares are held in nominee name versus record name.
A significant trend over the past five decades is the change from primarily certificated share ownership to nominee-held shares. Fifty years ago, when investors purchased stock in a public company, in many cases stock certificates were issued and the shares were registered directly with the issuer’s transfer agent. As shareholders became more active investors, and trade settlement changed from seven to three days, shareholders began to have their shares held in custody by a bank or broker to facilitate trading. Custodian- held shares are recorded in nominee name under the bank or broker holding the shares and are often called Street-held shares. The identity of the beneficial owners of these Street-held shares in many cases is not known to the issuing company. By 1990, the typical publicly-traded company had approximately 85 percent of its issued and outstanding common stock held in Street name.
Until 1980, there was little formal guidance for estimating carryover basis in B reorganizations, other than through direct shareholder inquiry (i.e., survey) or from the Target Company’s books and records. Some companies adopted the “Cohan Rule,” which provided the Taxpayer with a precedent in the event it did not have fully supportable basis data. In the absence of actual knowledge, an estimate of basis that was greater than zero (under the assumption that the stock price traded at greater than zero) could be considered reasonable.2
Seeking to formalize the basis determination process for B reorganizations and recognizing the burden of surveying 100 percent of the shareholders of a publicly traded company, in 1981 the IRS issued guidelines for determining target tax basis in B reorganizations. These guidelines, set forth in Rev. Proc. 81-70, primarily focused on using statistical sampling procedures to survey a subset of the Target Company shareholder base and extrapolating the basis data obtained from the sample group to all Company T’s outstanding shares. The guidelines prescribed that the resulting basis estimate was to achieve a sample error of 10 percent (or less) at a 95-percent confidence interval. Rev. Proc. 81-70 also included procedures for estimating target stock basis in the absence of actual knowledge (i.e., non-responding sampled shareholders). For example, the Acquiring Company could look at the date a stock certificate was issued to the Target Company shareholder and assign the average trading price on the date of issuance as the estimated basis.
To sample and survey a public company’s shareholders, you would first need to identify the shareholders and then randomly select a sample group of holders, both in Record and Street name. These steps may prove challenging because the majority of the publicly-held shares are held in Street name, which means the identities of a company’s shareholders are not readily apparent and oftentimes not available. Bank and broker nominees will often not divulge shareholder account information to the Acquiring Company, let alone shareholder basis data, out of a fiduciary responsibility to their account holders.3 Consequently, the Acquiring Company in some cases will encounter difficulty with ensuring statistical validity and obtaining sufficient Company T shareholder basis data to estimate the Company T shareholder’s aggregate basis within the confines of the guidelines set forth in Rev. Proc. 81-70.
On June 22, 2004, the IRS issued a request for comment in Notice 2004-44 on basis determination methods, partly because of concern that changes in the marketplace (i.e., increasing levels of nominee-held shares) made adherence to Rev. Proc. 81-70 problematic. 4 Some taxpayer advisers found it difficult to obtain sufficient survey response to produce statistically reliable basis results and, accordingly, turned to alternative basis estimation methods. The unique estimation methods incorporating stock trading models and SEC-filing data used by third-party advisers apparently led the IRS to reach out to taxpayers and their advisers for input on whether a change in basis estimation guidelines was necessary.
After considering the comments submitted and meeting with various representatives within the investment community, including the Depository Trust and Clearing Corporation (DTCC) — the clearinghouse for virtually all Street-held shares), the IRS issued Notice 2009-04, on December 12, 2008. Notice 2009-04 provides alternative methods for basis determination, incorporating many suggestions provided by taxpayers and their advisers, and reports the planned issuance of an expanded revenue procedure in the near future. Notice 2009-04 was issued as a safe harbor, providing the taxpayer a two-year window for determining target basis in B reorganizations and for adjusting basis computed under previously- used (not authorized) methods to conform to the newly issued guidelines. The alternative methods in Notice 2009-04 included using DTCC security position reports — lists of security holdings reported by each bank and broker having a position in a given security on a given day — and SEC-reported ownership positions (i.e., Series 13 reports) to estimate carryover basis. While the IRS apparently recognized the challenge of adhering to sampling guidelines with the provision of alternative basis estimate methods, Notice 2009-04 reaffirmed that the optimal method for determining the tax basis of Target Company stock was through the statistical sampling methods outlined in Rev. Proc. Proc. 81-70.
Under the alternative basis determination methods detailed in Notice 09-04, the IRS introduced certain guidelines that the Acquiring Company can use to derive a basis estimate by analyzing security position DTCC registration reports. By evaluating the change in historical shareholder positions (i.e. observing when shares are purchased and sold) and assigning basis to the share purchases, the taxpayer can estimate potential shareholder basis. Notice 2009-04 details three methods for determining how share sales are to be treated, ultimately affecting how shareholder basis is computed: (1) basis is reduced by share sales pulled from earliest share purchases (first-in-first out, or FIFO), (2) basis is reduced by share sales pulled from the latest share purchases (last-in-first-out, or LIFO), and (3) basis is reduced at historical average cost (AVG COST). Pursuant to Notice 2009-04, when using the alternative basis estimation approach, the Acquiring Company must adopt a basis estimate method — whether FIFO, LIFO, or AVG COST — that resulted in a lower basis. In other words, the taxpayer could elect which alternative procedure to use as long as it resulted in the lowest basis result.
2011 Expanded Revenue Procedure
Just over two years after Notice 2009-04 was released, the IRS issued Rev. Proc. 2011-35. Effective for transactions on or after June 20, 2011, the revenue procedure specifies four methods for determining carryover basis in basis transfer transactions. An overarching theme throughout the new revenue procedure is the potential for significant basis adjustment owing to missing or incomplete shareholder records or large sampling errors. Clearly, the burden rests on the Acquiring Company to provide adequate support for the target stock basis using one or more of the new revenue procedure’s methods. If the taxpayer cannot or does not use the methodologies prescribed in Rev. Proc. 2011-35, then section 3.01 of the procedure provides that basis in acquired Target shares may be established by such other methodologies as agreed by the IRS and Acquiring Company. Furthermore, taxpayers may use this revenue procedure retrospectively, for transactions before June 20, 2011.
The four methods prescribed by Rev. Proc. 2011-35 include certain guidelines from Rev. Proc. 81-70 and a modification of procedures introduced in Notice 2009-04. Rev. Proc. 2011-35 incorporates the statistical sampling process from the 1981 procedure with an important update to statistical precision guidelines. The shareholder survey method, another process for basis determination, is primarily a restatement of the guidelines included in Rev. Proc. 81-70. The two remaining methods are alternative basis estimation procedures which borrow certain techniques introduced in Notice 09-04. These alternative basis estimation methods utilize Form 13-F data, DTCC security position reports, and the transfer agent certificated detail list (also called “Master List”). In addition to these methods, section 3.02(3) reintroduces the term “Reporting Shareholders” from Notice 09-04, which is defined as a registered holder or nominee representing at least five percent of the vote or value of all outstanding Target shares, or an officer or director of the Target Company, or a plan that acquired Target stock for or on behalf of Target employees.
The Shareholder Survey method (Method No. 1) dictates the survey procedures that the taxpayer must use to obtain basis data, including the number of mailings and the mailing method. Method No. 1 is used whether the Acquiring Company opts to use the statistical sampling method or alternative basis-modeling methods. Pursuant to Rev. Proc. 2011-35, the taxpayer will need to survey all Reporting Shareholders, requesting basis information as well as relevant nominee details. The taxpayer must be prepared to provide support that a survey was indeed sent to a shareholder, adequate follow-up was conducted, and, if basis was secured, record of the basis provided and representation that the basis is reasonable (i.e., basis did not appear to be inaccurate as compared to the trading price of the stock during the period the shares were acquired). If the taxpayer does not make the required shareholder inquiry, the taxpayer may not utilize an alternative basis estimate. In other words, if the Acquiring Company does not demonstrate the effort to survey the shareholders as required, the Acquiring Company may not include that shareholder’s shares (or its basis) in the basis computation.
The statistical sampling approach (Method No. 2) remains relatively intact from Rev. Proc. 81-70 except for the expanded relative precision (sample error) guidelines. If the sample error does not exceed 10 percent, the taxpayer can use the basis estimate derived from the sample group. If the sample error is greater than 10 percent and less than 15 percent, the taxpayer’s allowable basis is formulaically determined and between the lower-bound (or the lower limit at the 95-percent one-sided confidence limit) and the sample basis estimate. If the sample error is greater than 15 percent, the taxpayer is required to submit as the basis estimate the lower bound at the 95-percent confidence limit. Such a defined basis adjustment had not previously been included in basis estimate statistical guidelines. Under the Registered Holder basis estimation method (Method No. 3), the taxpayer may estimate the basis for the registered shareholders using the certificate issuance information provided by the transfer agent. This method is not new to basis estimation techniques. Under Rev. Proc. 81-70, however, the taxpayer could use certificate issuance details to estimate basis only if basis was not obtained through other sources (such as a survey of Target Company shareholders or Target Company books and records). Method No. 3 essentially uses share amount and date of certificate issuance in conjunction with historical trading data to estimate registered holder basis. Under Rev. Proc. 2011-35, the taxpayer may elect to use Method No. 3 for all registered holders — excluding Reporting Shareholders. In addition, the transfer agent may have details on target shares that are recorded in bookentry form. Book entry shares are typically shares issued through a dividend reinvestment plan, and information on book-entry issued shares often includes share amount and date of issuance. Thus, the taxpayer may use basis estimation Method No. 3 for certificated as well as book entry shares.
The fourth and arguably most controversial of the basis estimation methods in Rev. Proc. 2011-35 allows taxpayers to estimate the tax basis for Street-held shares, or shares that are held in nominee name on behalf of the beneficial owner, rather than in Record name. Under Method No. 4, the taxpayer has two options for estimating Street-held shares: one incorporates security position reports issued by DTCC,5 and the other option incorporates SEC Form 13-F reported holdings.6 Each approach (SPR or 13-F) carries its own challenges. Both approaches, however, utilize historical average cost (AVG COST) as the method for computing estimated basis. (Rev. Proc. 2011-35 provides a particularly illustrative example of the mathematics behind the average cost calculation.) Use of average cost is a welcome change from previous guidelines. Under Notice 2009- 04, the taxpayer could compute estimated basis using FIFO, LIFO or AVG COST approaches, whichever method produced the lowest basis amount. The change to basis estimation procedures using AVG COST (alone) signals a more equitable approach, which arguably offsets the stiffer data retention guidelines discussed below.
In Notice 2009-04, the IRS raised the notion of estimating shareholder basis using DTCC security position reports (SPRs). Specifically, periodic SPRs were used to assess the buying activity of beneficial owners, with bank and broker nominee positions as the proxy for beneficial owner holdings. For example, Beneficial Owner A (BO A) and Beneficial Owner B (BO B) may hold 10 and 20 shares of T stock, respectively, with Broker X, as of Date 0. Broker X reports to DTCC its position of 30 shares in T stock (10 + 20 shares) on Date 0. Then, BO A and BO B purchase 5 and 8 shares of T stock, respectively, on Date 1. On Date 1, Broker X reports to DTCC 43 shares held in T stock (30 + 8 shares). The DTCC security position report (SPR) for Date 0 would indicate 30 shares in T stock held by Broker X, and the SPR for Date 1 would indicate Broker X held 43 shares in T stock. Under Notice 2009-04, the weighted average trading price between Date 0 and Date 1 can be assigned as the basis for the purchased 13 shares of T stock. To amplify this example, the group is expanded to 200 custodians (a typical SPR will have more than 200 bank and broker custodian positions) and the date range is expanded to include several weeks or months. By evaluating the increases and decreases of nominee (broker/bank) positions and assigning a potential price paid for new share positions, the taxpayer can estimate the potential basis of shares purchased in the market.
The SPR approach under Rev. Proc. 2011-35 (Method No. 4(i)) further hones the basis estimation process introduced in Notice 09- 04, adding stricter data collection requirements. The SPR approach requires the analysis of weekly changes in nominee positions in the Target stock for seven years, ending on the date of transaction/ merger, and assigning as basis the adjusted weighted average trading price for the period a purchase is made. If a Target Company has traded publicly for less than seven years, the SPR analysis defaults to the amount of time the company has been public. The seven-year period of analysis coincides with DTCC’s seven-year retention period for SPRs; DTCC operates under a trailing seven-year retention policy, where historical SPRs are continually removed from their databases. In total, the Acquiring Company would need to capture 364 weekly SPRs covering the seven-year period of analysis (52 x 7) when using the SPR approach to estimate shareholder basis for Street-held shares.
When adopting the SPR approach, the Acquiring Company must be diligent in retaining all 364 weekly SPRs. Pursuant to section 4.04 of Rev. Proc. 2011-35, for every SPR not captured, the taxpayer will receive a 1/364 basis adjustment.7 For example, the Acquiring Company retains three quarters of the required SPRs (i.e., 273 weekly SPRs) for T stock, and the estimated per share basis for all Streetheld shares in T stock is $10.00. The Acquiring Company must take a 25 percent downward adjustment for the 25 percent missing SPRs ((364-273)/364 = 25%), or a reduction of $2.50, for a final basis per share of $7.50 for all Street-held shares. Consequently, while the SPR approach is entirely data driven, requiring virtually no interaction with holders besides surveying Reporting Shareholders, the taxpayer needs to be vigilant with its data retention practices.
The 13-F approach (Method No. 4 (ii)), another procedure for estimating shareholder basis for Street-held shares, carries its own caveats. From a data retention view, the approach is more simplistic. The 13-F approach requires a review of 40 quarters of institutional ownership history — or 10 years of data. There are several venders for historical 13-F data and such records can be found as far back as 1988. As with the SPR approach, the 13-F approach involves evaluating changes in institutional holdings over time; here an increase in a quarterly 13-F reported position is interpreted as a purchase and a decrease in a quarterly position is interpreted as a sale. The basis assigned to purchases is the adjusted weighted average trad-ing price for the quarter the purchase is made, and the overall basis estimate is computed using the AVG COST approach.
Form 13-F filing institutions, however, constitute a subset of Street-held shares. Street-held shares typically include institutional investors, retail holders, short-term investors, and 401Ks and trusts. Consequently, an analysis of 13-F institutional-filing positions alone would exclude certain shareholders. The IRS acknowledges that institutions do not represent the entire Street community. To this end, the revenue procedure imposes a 25-percent basis adjustment to basis estimated under the 13-F approach. For example, the Acquiring Company utilizes the 13-F approach to estimate T stock and evaluates 40 quarters of institutional ownership data. The estimated per share basis for all institutionally managed (Form 13-F filed) shares in T stock is $10. To estimate basis for all Street held shares under the 13-F approach, the Acquiring Company must multiply the basis per share by 75 percent, resulting in a basis per share for all Street-held T stock of $7.50.
Two additional issues raised in Rev. Proc. 2011-35 are the timeliness requirement and the eligibility of Rev. Proc. 2011-35 basis study for the IRS Pre-Filing Agreement Program. In accordance with sections 1.351-3 and 1.368-3 of the revenue procedure, the taxpayer must file certain transferor information with its tax return for the tax year of the transfer basis transaction. While a statement can be included with the transaction-year tax return that a basis study is pending, to be in compliance with the guidelines the taxpayer must include complete statements including transferee’s total basis in the transferred stock with the return for the tax year no later than two years following the transaction. Moreover, to be in compliance with Rev. Proc. 2011-35, surveys sent to shareholders must be “substantially” completed within two years of the transaction. For transactions completed prior to June 20, 2011 (the effective date of Rev. Proc. 2011-35), a survey is considered timely and reporting requirements considered satisfied if they are completed by June 20, 2013, if the taxpayer adopts Rev. Proc. 2011-35.
The taxpayer does have an opportunity to modify the methods outlined in Rev. Proc. 2011-35. Under section 3.01, “[i]f a taxpayer cannot or does not use the methodologies prescribed in this revenue procedure, basis in acquired Target shares may be established by such other methodologies as agreed by the Service and Acquiring.” This provision sets the stage for the taxpayer to apply for the Pre- Filing Agreement program. In effect, section 6 of Rev. Proc. 2011-35 states that whether a study is completed in compliance with the procedures outlined may be the subject of a pre-filing agreement.8 This provides the taxpayer, if it desires, an opportunity to potentially modify the stringent guidelines in Rev. Proc. 2011-35.
The most important recommendation for a company anticipating a basis-transfer transaction is to capture shareowner information on the Target Company as soon as the deal is publicly announced. Conducting shareholder inquiries before a transaction closes allows sufficient time for the Acquirer to secure the relevant data and to notify data vendors of potential future requests for shareholder information. Inquiries may include: What is the ownership history of Target’s major insiders? What employee plans are in place and how is basis data retained? And what information is available on the Target Company’s historical Street-holder positions (i.e., how many SPRs have been captured)?
Overall, the three main methods for the Acquiring Company to consider are the following: (1) statistical sampling (Method No. 2), (2) SPR approach for estimating Street-held shares (Method No. 4(i)), and (3) Form 13-F approach for estimating Street-held shares (Method No. 4(ii)). Of the remaining methods, using only the survey method is most practical when the Target Company is closely held. The survey method (in part) is already a fundamental component of each of the main basis estimation methods, as the taxpayer is required to survey Target Company’s Reporting shareholders. Separately, the registered holder basis estimation method (Method No. 3) is a fairly straightforward process if the data are available from the transfer agent. Most transfer agents will retain certificated detail lists on acquired companies for a year or so. Estimating basis for thousands of registered holders, however, may be cumbersome, even if the data is in electronic form. The taxpayer may want to consider the value of estimating just a portion of the registered holder group.
First to consider of the main basis estimation methods is the SPR approach, with its significant data requirements. As of December 15, 2011, DTCC’s fee for each security position report is $120; therefore, to obtain 364 weekly SPRs could require an expenditure of more than $40,000. In addition, while two years of DTCC SPRs may be available in spreadsheet format, the remaining five years of older DTCC data are typically delivered in hard copy format, in a layout that may challenge most scanning software packages. That said, the key caveat to using the SPR approach of the Streetheld share basis estimation method, is the potential for a significant basis haircut for missing SPRs. For every week the Acquiring Company delays capturing historical SPRs from DTCC, one fewer weekly SPR could be available. To emphasize this point, Rev. Proc. 2011-35 announces that the IRS reserves the right to make “appropriate” adjustments where it determines that “SPRs not included in the [basis] determination represent material omissions.” Notwithstanding these stringent data requirements, the SPR approach could be the most straightforward and time efficient method for estimating Street-held shares, if all SPR data has been collected. The Acquiring Company may want to utilize the Street-held share basis estimation method 13-F approach if the transaction is more than a year-and-a-half old or less than three quarters of SPRs are captured. Consider this scenario: the Acquiring Company delayed capturing SPR data for one and three quarters years; i.e. 91 weeks of data are missing. This delay could potentially create a 25-percent reduction in basis (91/364 = 25%). Consequently, the Acquiring Company may be better off utilizing the Form 13-F approach, which would yield the same basis 25-percent adjustment as under the SPR approach.
The statistical sampling method, perhaps the more complex of the basis estimation methods, is generally more successful when the Acquiring Company conducts the study within a few years of the transaction. A successful statistical sampling study could be defined as one where the relative precision of the sample basis estimate falls at or below 10 percent. Under such circumstances, the Acquiring Company may use the sample basis estimate per share as the per share basis estimate for all outstanding shares (exclud-ing Reporting Shareholder shares). Precision is typically improved with greater sample representation. Surveying Target Company shareholders soon after the transaction closes may increase shareholders’ willingness and ability to respond with basis data. Furthermore, bank and broker custodian account data may also be more accessible when queried closer to the transaction. Nominees face data retention policies where historical account data may be moved off active data servers. And under certain circumstances, bank and broker nominees may be able to provide shareholder basis data directly to the taxpayer, avoiding the need to survey the shareholder entirely. If timed well, and with the right insight into the Street community, the statistical sampling method could provide a basis estimate less likely to receive an adjustment.
Overall, identifying which basis estimation method best meets the needs of the Acquiring Company is a function of study timeliness, data availability, sensitivity to basis adjustments and budget pressure.
Juliane Keppler is a Managing Director within NASDAQ OMX Corporate Solutions Investor Analytics team, assisting clients with shareholder identification and analysis for regulatory compliance. Ms. Keppler has been in the target stock tax basis determination business for 20 years, with the past 15 years at a Big Four accounting firm. She received her B.A. degree from Clark University and M.B.A. (Finance) from City University of New York (Baruch). Her previous article on stock basis determination guidelines appeared in the July-August 2004 issue of The Tax Executive. She may be contacted at Juliane.Keppler@nasdaqomx.com.
1. FactSet Mergers.
2. The Cohan Rule is named after George M. Cohan, the songwriter, who claimed on his taxes certain travel and entertainment expenses during years 1921 and 1922, without the benefit of receipt documentation. The IRS disallowed the expenses. Mr. Cohen argued in court that, although he did not retain the entertainment expense receipts, there was in fact an amount spent and the amount spent was greater than zero. The court agreed, setting a precedent for estimation in the absence of actual knowledge. See Cohan v. Commissioner, 39F. 2d 540 (2d Cir.1930).
3. Acquiring companies can obtain a list of non-objecting beneficial owners, which would provide name, address, and share position, and is typically available for six months after a deal closes. NASDAQ OMX’s Investor Analytics team has found that NOBO information can cover between 25 to 75 percent of Street-held shares, with higher levels of retail Street-held shares typically driving up the proportion of NOBOheld shares.
4. Juliane Keppler, “Basis Studies Are Given Red Flags — Revenue Procedure 81-70: Past, Present and Future,” 56 The Tax Executive 202 (July-August 2004).
5. Under section 3.02, the term SPR also includes similar publications for other clearinghouses, whether domestic or foreign.
6. Form 13-Fs are the SEC report filed by institutional money managers managing an excess of $100 million in assets and are required to be filed quarterly, listing the security positions under management.
7. Rev. Proc. 2011-35, § 4.04 (2)(d)(3)(i).