Basis studies are given red flags Revenue Procedure 81-70: past, present and future.

Taxpayers take heed if your company has been involved in a tax-free stock-for-stock acquisition. Has tax basis of the acquired stock been determined? The Internal Revenue Service is scrutinizing extant tax basis studies and closely reviewing the continuing vitality of a two decades old revenue procedure on basis determination.

If your company engaged in a tax-free stock acquisition and has already computed the tax basis of the acquired stock or believe that in the future you may need to do so, you should read on. This article reviews the history of Revenue Procedure 81-70, which addresses the process for determining carryover basis of widely held acquired stock, including the expansion of its application in the 1990s. It also explores the auditing of tax provisions under the Sarbanes-Oxley Act, which has shone a brighter spotlight on basis studies and accentuated the need for the IRS to revisit its scope and effectiveness. Finally, the article concludes with a discussion of the IRS's recent request for comment and the various possible options available to the taxpayer.

How It Began

Corporate acquirers have long been able to use their stock to acquire the stock of another company (target). 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. In a B reorganization, the acquirer's basis in the newly acquired target stock is carryover basis. (I.R.C. [section] 362(b).) In other words, the target shareholders' basis carries over to the acquiring corporation.

For example, assume company T has 5 stockholders, with 10 shares each and a tax basis of $10 per share--for a total tax basis of $500. If company A uses A voting stock to acquire all 50 shares of company T in a B reorganization, then company A's tax basis in the T stock will be $500, the aggregate of the shareholder's tax basis in the T stock.

If the acquirer knows the identity of the target's shareholders, the collection of this tax basis data can be very straightforward. The identification process becomes more problematic, however, if the acquired company is publicly held because many shareholders choose to have their shares held by a nominee instead of in certificated form. The nominee could either be a bank or a broker holding custody of the shares on behalf of the shareholder in their account. Obtaining basis information from each of the target's shareholders now becomes a more administratively burdensome task.

In 1981, recognizing the extensive undertaking of retrieving tax basis information from thousands of shareholders, the IRS issued a revenue procedure affording taxpayers an alternative to surveying each of the target shareholders. Revenue Procedure 81-70, 1981-2 C.B. 729, was introduced to provide the taxpayer with choices; that is to say, it was essentially a relief procedure.

Rev. Proc. 81-70 recognizes that in the event of a widely held target company the potential exists for shareholders to not respond to a tax basis inquiry. This would make determining carryover tax basis essentially impossible. Thus, the revenue procedure permits taxpayers to use statistical sampling methods to identify a sample group of shareholders from the target stockholder population. The revenue procedure also permits the use of tax basis estimation through stock certificate analysis, in the absence of known actual shareholder basis.

The statistical methodology the IRS adopted in Rev. Proc. 81-70 followed commonly accepted statistical analysis rules. Specifically, the level of probability that the resulting overall basis estimate would fall within a given confidence interval should be 95 percent. The sample error, in relationship to that overall basis estimate, should be no greater than 10 percent.

The revenue procedure also provides guidance on what the taxpayer should do in the event of non-respondents to a tax basis inquiry. Here, the procedure refers to the registered holders list maintained by the transfer agent, who provides details of stock certificate issuance and cancellation. In the absence of actual shareholder tax basis information, the procedure recommends that the taxpayer use the certificate issuance date and corresponding trading price to assign stock tax basis.

While Rev. Proc. 81-70 gives a taxpayer alternatives for determining carryover tax basis in "B" reorganizations, the full scope of this revenue procedure may not have been fully appreciated until another form of tax-free acquisition fell under its jurisdiction. In 1995, Rev. Proc. 81-70's use was expanded to apply to certain reverse triangular mergers under Treas. Reg. [section] 1.358-6(c)(2)(C) (ii) (B). These are so-called reverse triangular merger/"B" overlap transactions--compare section 368(a)(2)(E) with section 368(a)(1)(B). Here, the acquiring company creates a transitory subsidiary that is "merged with and into" the target company, with the target company surviving after the merger close. The sole consideration issued to the shareholders of the target company is voting stock of acquiring company.

The importance of Rev. Proc. 81-70 increased along with the use of stock as consideration for company acquisitions. In the late 1990s, with heightened acquisition activity and a rising stock market, company stock was often the consideration in corporate acquisitions. During the five-year period from 1997 to 2002, for example, more than 2,500 corporate mergers used stock as the sole consideration for the transaction. Acquirers needed to know the carryover basis for these acquired companies. Hence, the need for Rev. Proc. 81-70 basis studies increased.

Where We Are Now

As discussed in "The IRS Reviews Revenue Procedure 81-70: The Stakes for Taxpayers Are High," 55 The Tax Executive 122 (March-April 2003), the IRS became concerned that basis studies were being done in ways that did not fully comply with the revenue procedure. Some studies placed a significant emphasis on basis data derived directly from target companies' shareholders, whereas other studies placed a greater emphasis on using stock trading assumptions for the purpose of estimating carryover basis.

The question is how does the taxpayer obtain the target shareholder tax basis data or, more important, who has the basis data. A publicly held target company could have several layers of ownership, from certificated shareholders to shares held in custody by a myriad of bank or broker custodians. Gathering basis information from listed registered holders is straightforward. Complexity emerges when you seek the tax basis of shares held in "street" name by pension plans, 401K accounts, mutual or hedge funds, private portfolio management, and non-US entities.

The disparity in basis determination approaches being used in stock basis studies reflected a tension between the feasibility of obtaining basis information directly from "street" holders and the assertion that estimating the tax basis of their share positions was the only practical option. Stated bluntly, some providers of basis study services had not experienced significant difficulty gathering the basis information needed to successfully complete a basis study in compliance with Rev. Proc. 81-70. Other providers did establish the need to use methods not provided for in the revenue procedure, such as trading behavior assumptions and SEC reporting documentation pertaining to institutional money management to compute an overall basis estimate.

The IRS's conundrum: Which is right? How much interpretative latitude should the IRS allow? Can the agency accept an overall tax basis figure on face value with minimal support documentation? The stakes are high and the magnitude of the issue is not lost on the IRS. There have been many stock deals with values in excess of $1 billion.

Late in 2003, the IRS decided to evaluate both the prevalence of basis studies and the range of possible approaches to refining basis studies. Thus, the IRS developed for use in ongoing large case audits an information document request (IDR) on taxpayer completion of a Rev. Proc. 81-70 study during the audit cycle in question. This approached gave the IRS an opportunity to review studies prepared by various third party providers.

Here is what the IRS found: Some of the basis studies do not comport fully Rev. Proc. 81-70, either because they did not comply with the letter of the revenue procedure or because the taxpayer's documentation did not support tax basis estimation claims.

Following the enactment of the Sarbanes-Oxley Act, taxpayers have a clearer need to scrutinize external basis studies rather than taking them at face value. Thus, the 2002 legislation effectively requires taxpayers to institute (and document) sophisticated internal controls over financial reporting. Consequently, reports conducted by third party providers are more likely to be internally audited for accuracy and adherence to IRS regulations and procedures, whereas prior to 2002, taxpayers may not have questioned such provided computations.

On June 22, 2004, the IRS issued a request for comment in Notice 2004-44. The notice specifically states that the IRS is troubled that changes in the marketplace may have made adherence to Rev. Proc. 81-70 "impossible" and, therefore, taxpayers are not complying with its requirements. The notice also mentions that commentators have suggested Rev. Proc. 81-70 needs to be revised. (Interested parties have until September 30, 2004, to submit their comments.)

Where Do We Go from Here

The IRS typically demands supportable accuracy and statistical reliability. Some third party providers say the level of acceptability has to be relaxed. Either way, the taxpayers need a compromise on approach procedure or, at a minimum, some clarity.

Conducting a statistically random sample of a large population may in and of itself provide relief to the taxpayer. In doing this, the taxpayer does not need to obtain tax basis data from all its shareholders. What Rev. Proc. 81-70 does require is that taxpayers show a best effort in collecting actual basis data.

How do you define best effort? At what point can the taxpayer adopt alternative methods for determining carryover basis, thereby bypassing the use of actual basis data in the overall carryover basis computation? Some say that alternative methods for estimation can be used when the confidence level (sample error) falls outside the 10-percent range, once a substantial amount of actual basis has been obtained. Further, and more important, taxpayers need clarity on what is an acceptable method for estimating basis. Should the taxpayer use trading history alone? Which share basis valuation approach should be used--first-in-first-out or last-in-first-out? How much latitude does the taxpayer have if statistical sampling is not used? What if the shareholders have not all been documented?

The timeliness of the Rev. Proc. 81-70 basis study may answer some of these questions. With every passing year, historical shareholder information is lost. As information ages, it is archived or--worse--discarded. In fact, the Depository Trust Company, the repository for information on "street"-held share positions, removes data from its computer system after seven years. The taxpayer can benefit substantially by capturing as much current shareholder data as possible. In other words, the taxpayer is best protected by requesting the tax basis of the targets' recently turned-over shares as soon after the close of transaction as possible.

What is not widely known--or perhaps it is but is not adhered to--is that carryover basis is required to be included as part of the return for the tax year in which the reorganization occurred. (Treas. Reg. [section] 1.368-3(a)(2); see also Rev. Proc. 81-70, [paragraph]1, [section] 2.) Since this is both required by regulation and mentioned in the revenue procedure, the IRS finds unacceptable, and is less sympathetic to, the taxpayer's response of data unavailability years after a transaction closes.

The challenge for the tax director is convincing senior management that the money spent on a basis study today may prevent significant adjustments for future non-compliance. This issue of timeliness adds yet another wrinkle to the IRS's Rev. Proc. 81-70-compliance challenge.

Some feel the IRS should give credit to taxpayers who comply with the procedure's timeliness guidelines, by conducting a Rev. Proc. 81-70 basis study within a reasonable period following the transaction close. Further, credit should be given to taxpayers who obtain a majority of the needed actual basis information directly from shareholder records. Here a best effort of adherence to the spirit of the carryover basis rule would be shown. On the other hand, the IRS would likely be less receptive to a basis study, conducted several years after the transaction, which is based predominantly on assumptions.

A two-tiered approach may be worthwhile. For those taxpayers, and their third party providers, who make the effort to comply either with timeliness or through actual basis information--provided statistical reliability is assured --the IRS could favorably view the overall basis with little or no adjustment. For those taxpayers, and their third party providers, who choose to use a more assumptions-based approach, the IRS could accept the overall basis but only after a sample-error haircut, adjusting for the additional risk assumed by not using actual basis data.

How the IRS proceeds may be a matter of expediency. While published guidance may achieve public awareness going forward, it may not achieve immediate effectiveness for what may be many basis studies waiting to be reviewed. The IRS may simply issue a directors' memorandum, outlining how the agency will allocate resources with respect to targeting egregious non-compliance.

Taxpayers previously involved in tax-free stock-for-stock exchanges qualifying as either a "B" reorganization or a reverse triangular merger could be significantly affected by the IRS's final word on basis study computation. If a Rev. Proc. 81-70 basis study was conducted with more assumption-based calculations than actual basis figures, the taxpayer may find itself faced with an adjustment in computed tax basis. Moreover, if a basis study was not conducted contemporaneously with the transaction close, a more thorough basis gathering technique could be in order. The greater the percentage of basis data derived from actual shareholder responses, the more reliable the basis calculation and the less vulnerable the taxpayer.

JULIANE LAURA KEPPLER is Managing Director of Shareholder Services, a division of KPMG's Mergers and Acquisitions practice, where she conducts stock basis studies to assist clients in complying with IRS reporting requirements. She received her M.B.A. degree in 1991 and has been specializing in shareholder demographics for more than 15 years. Her previous article on Rev. Proc. 81-70 appeared in the March-April 2003 issue of The Tax Executive.