Part II   Accounting Firms

4   The Skunk Works

The emergence of KPMG paralleled that of the other major accounting firms. Its predecessor firm, Peat Marwick, was founded in 1911 and specialized in providing audit and tax services to banking and financial institutions. During the 1950s, the firm was the most aggressive and innovative of the Big Eight, emerging as the largest and highest grossing accounting firm in the United States.¹ The firm grew by rapidly buying up small firms and expanding its consulting services. In 1987, Peat Marwick joined forces with a second-tier firm to become KPMG Peat Marwick, at the time the biggest accounting firm in the world.² This was the first in a series of consolidations that turned the Big Eight into the Big Six by the late 1990s.³ (The merger between Price Waterhouse and Coopers & Lybrand in 1998 then brought the number down to five; Arthur Andersen’s bankruptcy in 2001, to four.) Over the next ten years, tax services continued to expand at KPMG.⁴

During most of its history, the firm had been organized according to different geographic areas. Under this structure, tax professionals in a particular office reported to the tax partner in charge of that office, who in turn reported to a tax partner higher up in the organizational hierarchy. In the early 1990s, the firm was reorganized along industry groups. After the reorganization, tax professionals reported to tax partners within their industry group. The reorganization increased each group’s capacity and incentives to develop specialized services to suit the needs of clients in particular sectors. It also facilitated the capacity to cross-sell services and products through audit services that served the same industry. The reorganization gave rise to new performance measures under which revenue goals were established for different industry groups. Revenue targets had earlier been deemphasized, but now became an important focus of the tax practice.⁵ With the shift to industry groups, a tax professional’s compensation benefits and advancement became tied to the success of his or her group.

In the 1990s, tax leaders at the firm also began to rethink their fee structure. Like corporate law firms, accounting firms had traditionally charged for their services based on a billable hour. At some point in the 1990s, partners at KPMG became aware that competing firms had moved to value-based fees, typically calculated as a percentage of the tax savings obtained. This move could significantly increase the revenue potential of accounting firms.⁶ Under the leadership of John Lanning, who was named head of Operations Tax Services in 1996 and vice chair of Tax Services in early 1998, the firm began to emphasize designing tax products that could be sold to numerous similarly situated clients and priced on this basis. Jeff Stein, the partner in charge of International Tax Services and next in line to become vice chair after Lanning retired, was an equally enthusiastic proponent of generating fees by focusing on tax products.

“A Brilliant Tax Strategy”⁷

The tax product bug caught on quickly at KPMG. Spearheaded by partners in the Washington National Tax (WNT) office, the firm’s technical headquarters and think tank, several colleagues formed a group they called the “Skunk Works.”⁸ The group drew its name from the original Lockheed Corporation Skunk Works, which worked secretly to develop stealth fighter planes to deploy against the Axis powers during World War II.⁹ Members of KPMG’s Skunk Works met informally every few months to discuss ideas for tax strategies that would reduce or completely eliminate their clients’ tax liability.¹⁰ As tax scholar Calvin Johnson notes, “A ‘skunk works’ operation was once a secret research lab for developing planes to defeat the Nazis and the Communists. The KPMG tax skunk works dreamed up transactions against our United States.”¹¹

One tax stratagem that seized the group’s imagination involved a transaction designed by Seagram and Dupont, two Fortune 500 multinational companies, that had generated considerable attention a few years earlier. In April 1995, Seagram needed cash because it wanted to acquire MCA, Inc., a media company valued at several billion dollars. A source of potential cash was about 160 million shares of Dupont that Seagram owned, which Dupont was interested in buying back. If the transaction was treated as a sale, however, the proceeds would be taxable at a 35 percent rate. Seagram’s tax advisors designed a sophisticated transaction under which Seagram could, for tax purposes, treat the proceeds as the distribution of a dividend by Dupont. As a result, the proceeds of the transaction were taxed at an effective tax rate of 7 percent.

This aggressive tax avoidance transaction saved Seagram more than $1 billion in taxes and was widely reported in the business and tax press.¹² The transaction was the subject of heated debate in the tax bar, with several observers concluding that the tax savings, which were premised on a literal reading of the statutes and regulations, were legitimate. Shortly after the transaction became public, Congress amended the tax provision on which the tax benefits were premised.¹³

The KPMG Skunk Works members became excited when they read about the Dupont-Seagram strategy. By taking advantage of the interaction of two provisions of the Internal Revenue Code, Seagram’s ingenious transaction made a very substantial capital gain all but disappear for tax purposes. In 1995 Stan Wiseberg, a partner in KPMG’s WNT office, expert in corporate reorganizations, and convener of the Skunk Works, held freewheeling discussion sessions to develop a transaction that would take advantage of some of the same regulations to convert taxable into nontaxable gains.¹⁴ Several other WNT partners were active participants, including Joel Resnick, considered an elder statesman in the office, and Richard Smith, a partnership tax specialist who had been recruited from the IRS earlier that year.¹⁵ Another regular attendee was John Larson, a manager in KPMG’s San Francisco office and member of the International Services practice.

Shortly after the group began discussing a strategy based on the SeagramDupont ploy, Larson ran into David Lippman and Michael Schwartz, two other members of the International Services practice, at an internal KPMG tax conference. Lippman and Schwartz worked from KPMG’s New York office, where some of the most technically sophisticated tax professionals at the firm outside of WNT were located. In the course of their cocktail party conversation at the conference, Larson explained the idea for the strategy to Lippman and Schwartz. As the three colleagues spoke, they realized that the strategy might have huge potential. After the conference, the three continued to exchange ideas and drafts. Robert Pfaff, a partner in the Denver office and occasional participant in the Skunk Works, also began to provide regular suggestions and feedback.

The transaction designed by Lippman, Schwartz, and Larson exploited the same tax regulations as the Seagram-Dupont deal. The goal of the KPMG strategy, however, was not to structure an actual sale to minimize tax effects. Instead, the collaborators’ purpose was to design a transaction that created capital losses that an individual taxpayer could use as deductions to offset gains from an unrelated transaction. The target clients were entrepreneurs who had a one-time gain from the sale of their business and needed a loss to shelter that gain from taxation. In the mid-1990s, many business owners in the San Francisco, Denver, and New York areas fit this description. The goal of the transaction was to create a large artificial loss.

The strategy was intended to exploit a rule that applies in what are termed “incomplete redemptions.” This rule, reflected in a Treasury regulation, states that if an amount received in a redemption is treated as a dividend (as opposed to the proceeds of a sale) then “a proper adjustment of the basis of the remaining stock will be made.”¹⁶ The fundamental idea underlying this rule was that when the sale of stock back to the company is deemed to be a dividend distribution, the seller of the stock should not lose all of his or her basis in it—that is, the amount paid for the stock at purchase that would be deducted from the proceeds of the sale to determine taxable gain if the transaction were treated as a sale. In this way, the basis is shifted to any remaining stock owned by the shareholder.

A second set of rules provides that a party is entitled to this shift in basis can apply it to any property held by a “related party.” The Seagram-Dupont transaction partly relied on these rules. The rationale underlying these rules is that property held by related parties is attributed to both. Thus, for instance, a wife’s ownership of stock is attributed to her husband and vice versa. If one spouse has no stock to which to apply the basis, the basis can be applied to the stock of the other spouse.

Under the transaction designed by the Skunk Works team, the taxpayer who was seeking the loss would purportedly be able to take advantage of the basis shifted from a related party—in this case a foreign corporation. It was essential that the corporation not be subject to taxation either in the United States or in the country where it was incorporated. The strategy involved setting up a shell company incorporated in the Cayman Islands, a tax haven. The transaction was designed so that the basis that shifted to the taxpayer was the same amount as the loss she wanted to claim. That way, when the transaction was unwound, the taxpayer could claim a loss equivalent to the adjusted basis, which could be used to offset capital gains.

The Foreign Leveraged Investment Program or FLIP, as it came to be known, worked something like this: Assume the client had a $100 million gain she wanted to shelter. The shelter’s promoters would set up a corporation in the Cayman Islands; call it Cayman. The client would buy options to purchase 85 percent of Cayman at a premium. This payment was the mechanism through which the taxpayer paid her fees for the transaction, which were set at 7 percent of the tax loss desired. Having purchased these options, the client and Cayman were now “related”: The client was deemed to own Cayman under the ownership attribution rules in Section 318 of the Code—just as Seagram was deemed to own a part of Dupont by virtue of continuing to hold warrants to purchase Dupont stock after the Dupont stock redemption. Next, Cayman bought $100 million worth of a foreign bank stock, which it purchased with $100 million borrowed from that same bank. Meanwhile, the client bought a modest amount of the foreign bank stock.

The next steps were a little more complicated. The foreign bank redeemed its stock from Cayman at the same time that the client bought options to purchase the equivalent number of shares in the bank. Cayman used the proceeds of the redemption to pay back the foreign bank. In actuality, while there were ledger entries to indicate that the foreign bank loaned the money for the original stock sale to Cayman, the money never left the bank. Since Cayman and the client were supposedly related, the client’s options to purchase were constructively owned by Cayman. Cayman therefore had no reduction of its ownership interest in the foreign bank—just as Seagram’s ownership of warrants to buy Dupont prevented a reduction of its ownership interest in Dupont. And just as Dupont’s redemption of Seagram’s shares was treated as a dividend distribution to which no basis attached, so the foreign bank’s redemption of its shares was, supposedly, a dividend distribution to Cayman. Cayman was not especially concerned that the dividends would be taxable under U.S. law since the company was not subject to U.S. law, nor, for that matter, to tax laws in its home jurisdiction.

Having had all its shares redeemed by the foreign bank, Cayman had no remaining shares that could be adjusted to reflect the $100 million basis of the shares that were redeemed. The taxpayer, however, still had a small number of foreign bank shares. As a party related to Cayman, the taxpayer could apply that basis to these shares. In other words, the basis in the tiny amount of foreign bank stock that had been purchased was adjusted by $100 million. Now all the client needed to do was sell these shares at market price, and—voila—under a literal reading of the rules, the taxpayer could claim a $100 million loss. At the 1998 capital gains rate of 20 percent, the transaction according to its promoters enabled the client to avoid $20 million in taxes.

When the transaction finally came to light years later, tax scholars who reviewed it doubted that it met the technical requirements of the applicable tax provisions. The transaction was also invalid under several equitable doctrines, including the step transaction doctrine. Under this judicial doctrine, courts can determine the tax implications of a transaction by collapsing and then ignoring interim steps. As Johnson explains, “If we collapse the steps between [the foreign bank] and Cayman, there is no borrowing, no purchase of stock, and no redemption and repayment, only the $100 million staying in the [foreign bank] vaults. Nothing rode on the Cayman purchase and sale back of foreign bank stock, except the generation of a claimed $100 million loss that didn’t really happen.”¹⁷

The transaction, in addition, lacked economic substance. Although its developers tried to camouflage the absence of economic reality by adding a series of trading maneuvers, at bottom, the taxpayer did not have a “reasonable expectation” of pretax profit. Cayman typically held its foreign bank shares for two months. Having paid 7 percent for its options to buy Cayman, the taxpayer would have had to reap an annual return greater than 150 percent—not impossible, but certainly not reasonable to expect either.¹⁸ Unlike the Seagram-Dupont arrangement, the only reason for entering into the transaction was to generate a tax loss.

FLIP Goes to Market

As Lippman, Schwartz, and Larson developed the transaction, they realized they needed an investment advisory company to execute the steps of the transaction and bring in a bank to provide the loan. In May 1996, Larson got in touch with Jeff Greenstein, the principal of Quadra Investment, whom he knew had contacts at Union Bank of Switzerland (UBS). Greenstein was soon on board and a few months later approached the bank about participating in the transaction.

During the spring, the FLIP team also realized that a concurring opinion from an outside law firm would strengthen the marketability of their product. Bob Pfaff thought of R. J. Ruble, a young tax partner at the law firm Brown & Wood, whom he had heard give a talk. Beginning in the late 1990s, Ruble was a featured speaker at prestigious Practicing Law Institute lectures on the ethics of corporate tax lawyers. Tax professionals in the New York area regularly attended these sessions to comply with statewide bar requirements for continuing education in ethics. A self-described “tax shelter lawyer,” Ruble argued in one lecture that a tax lawyer’s responsibility in advising a client was solely to further the client’s interests. Tax lawyers, he insisted, had no separate obligation to the tax system. Decrying the “shrillness” of the public response to the tax shelter problem, Ruble reminded his audience that the tax system was an “adversary system.” Whatever problem existed, he contended, was the government’s fault for having drafted too many rules. “If we are able to use these rules to our clients’ benefit and the government’s detriment, we are merely doing our job,” he concluded.¹⁹ Pfaff liked Ruble’s approach.

Ruble discussed the strategy with Pfaff and the KPMG team and agreed to provide a concurring “independent” legal opinion to clients who purchased FLIP. He also began providing suggestions on the opinion letter the firm was drafting, that is, effectively writing the draft himself. Over the next year, he established an alliance between his firm and the tax products group at KPMG to “jointly develop and market tax products and jointly share in the fees.”²⁰ He also worked out an agreement with KPMG so that every time Brown & Wood’s name was mentioned during a sales pitch, the firm would earn a $50,000 fee regardless of whether it had provided an opinion letter in connection with the transaction.²¹

By fall 1996, KPMG had finalized a draft opinion letter concluding that FLIP, if challenged by the IRS, would “more likely than not” be upheld in court. Schwartz and other KPMG colleagues began to introduce clients to the strategy.²² The shelter’s designers also prepared a “pitch” book, entitled “Generating Capital Losses,” which explained the steps of the transaction.²³ FLIP was priced at 7 percent of the anticipated tax savings and was intended to be marketed to clients who were seeking a minimum $10 million loss.

The Perez brothers, from McAllen, Texas, were early purchasers of FLIP. Sons of migrant workers, the three brothers had developed a successful business providing ambulatory health services which they planned to sell to a large corporation. As they were finalizing the sale, they were contacted by a line partner in KPMG’s Dallas office. This partner had no prior contact with the family, but had obtained the Perez name from an advisor to the brothers who was subsequently paid a large finder’s fee. When they first heard about the strategy, the Perez brothers asked to consult with their lawyer, but were told that the strategy was proprietary. In a pattern that was repeated in subsequent FLIP cases, they were asked to—and did—sign a confidentiality agreement. They were also told that the strategy was legal and would not be audited by the IRS. As they discussed the strategy among themselves, one of the brothers observed: “I guess that’s how rich people do it.” Overcoming their initial reservations, the Perez brothers purchased the strategy.

The Perez transaction was not the only FLIP case where methods such as finder’s fees were used that raised questions about the independence of the advisors involved. First Union Bank, based in North Carolina, entered into an agreement with KPMG under which the bank provided marketing assistance for the firm’s tax products and earned $100,000 for each client referred. The close relationship that developed between the bank and the accounting firm mirrored the relationship between Carolyn Branan, a partner in KPMG’s Charlotte office and an enthusiastic promoter of FLIP,²⁴ and her husband, Ralph Lovejoy, who managed First Union’s Personal Financial Consulting Group. Neither Lovejoy nor Branan thought it necessary to inform clients of their personal connection, even though it might have raised a question about whether either was able to provide independent advice to the clients they shared.

With revenue from the first half dozen FLIP transactions flowing in, KPMG tax leaders began to appreciate the enormous profit potential of charging value-added fees for commoditized tax strategies to wealthy individuals. At the same time, they recognized that the tax product sales process was disorganized.²⁵ Line partners were using different versions of a client engagement letter, which did little to protect the firm against being sued if the strategy went bad. Occasionally too the strategy was implemented in a manner that made it difficult for the taxpayer to claim any tax benefits.²⁶

Recognizing that lack of centralization and organization could put the firm at significant legal risk, Lanning, the firm’s head of Tax Services Operations, created a new Department of Professional Practice for Tax (DPP) in February 1997.²⁷ Although such departments already existed in consulting and attest services, no firm-wide compliance and risk management system existed for tax strategies. Lanning asked Richard (“Larry”) DeLap to head the department. DeLap, who was based in Silicon Valley, had developed expertise in the taxation of technology start-ups and was the national partner in charge of the firm’s tax practice for high-tech companies after the firm’s reorganization in 1993. Within the firm, DeLap had a reputation as a thoughtful and independent professional. Although the newly created DPP was located in the WNT office in Washington, D.C., Lanning agreed to let DeLap run the group from KPMG’s Silicon Valley office.

The DPP’s first order of business was to review the draft opinion letter for FLIP, which had never been formally vetted by an independent group inside KPMG. The need for review was all the more pressing because in March of that year, the U.S. Tax Court issued a very important decision exhibiting newfound skepticism about tax shelters. In ACM Partnership v. Commissioner, the Tax Court in 1997 set aside a complex series of transactions engaged in by Colgate Palmolive to create a $98 million tax loss.²⁸ In ruling against the company, the court revived the business purpose doctrine, which in recent years had been somnambulant. Because the court concluded that there was no genuine business purpose served by the transactions other than to reduce taxes, it disallowed Colgate Palmolive’s claimed loss.

DeLap requested several tax partners by email to evaluate the technical merits of the strategy as well as how it was being marketed.²⁹ During a conference call in early April, the review group concluded that the firm could continue to market FLIP despite the ACM decision, but requested Pfaff and Larson to “beef up” the opinion letter. During the following months, DeLap tried to impose some modicum of order on the marketing and implementation process, insisting that line partners who were selling FLIP not deviate from the approved form letters, obtain appropriate waivers of liability from clients, and make sure to obtain signed engagement letters from clients before implementation.³⁰ All of these procedures were intended to protect the firm in what was recognized as “clearly a high-risk practice.”³¹

DeLap’s appointment to head the newly created DPP coincided with firm-wide efforts to emphasize products over individualized services. Under Lanning’s leadership, the firm committed significant resources to designing and marketing generic tax “solutions.”³² Eager to become the industry leader in this area, the firm established the Tax Innovation Center (TIC) inside the KPMG WNT office, the technical support center for its tax services nationwide. The TIC was charged with encouraging tax professionals throughout the firm to submit ideas for new tax strategies and shepherding the more promising ideas through the development process. To this end, the TIC awarded “light bulb” prizes—specially designed paperweights—and cash prizes to tax professionals who submitted ideas that were developed successfully.³³ To disseminate information about new tax products across the firm, the TIC regularly issued Tax Product Alerts that explained the new product, its “target” audience, its pricing structure, and the steps to be taken to deliver it. The notice also included access to a product “tool kit” that contained sample letters, a sample product, and a client “Q &A” that anticipated questions a prospective client might ask.

An early Tax Product Alert even went so far as to propose selling more-likely-than-not opinions as products to clients who were contemplating transactions that might be challenged as tax shelters. The 1997 Taxpayer Relief Act expanded the definition of tax shelters subject to underpayment penalties, but preserved a safe harbor for taxpayers who had reasonably relied on the well-reasoned opinion of a tax advisor in entering the transaction. With the broadening of potential taxpayer liability, the TIC apparently saw a new marketing opportunity. “Because of the tax law change,” it underscored, “this product potentially has a huge market and can generate high margin (and in many cases large fee) engagements when priced on a value-added fixed-fee basis.” Potential targets for the product were clients who planned to engage in transactions with “significant federal tax risk.” As a fee, the Tax Product Alert suggested 10 percent of the penalty exposure that the opinions purportedly prevented from being assessed.³⁴

To enhance its tax products focus, the firm engaged in a lawyer-hiring spree, recruiting aggressively at law schools and attracting partners away from traditional corporate law practices with multimillion-dollar deals.³⁵ One ad that ran in the ABA Journal depicted a rock climber rappelling down a sheer cliff face. In bold letters, it urged tax lawyers “to choose their own path” “in a changing landscape.” Clearly intended to lure young tax associates from traditional—read “staid”—corporate practice, the advertisement promised exciting work at the firm’s WNT office working alongside the “nation’s leading tax authorities.” It also offered the opportunity to work with the TIC, “ground zero for the identification, analysis, packaging and distribution of new high-growth tax products and services.”³⁶ In 1999, John Lanning, vice chair of Tax Services, a position that put him directly below the firm’s chair, applauded the firm’s success in recruiting tax lawyers, noting that the firm had hired 175 professionals with law degrees in the previous year, more than half of them right out of law school.³⁷

To obtain greater penetration among current clients and attract new ones, the firm hired a sales force that would focus exclusively on promoting tax products. In December 1998, Lanning announced the firm’s new “Tax Sales Organization” which, he predicted, would produce $200 million in revenue by the end of the year. Business Development Managers (BDMs), as they were called, were not recruited based on their tax expertise, but on their substantial sales experience. Paid a base salary and a substantial commission on each product sold, BDMs were expected to “hel[p] create an aggressive sales culture” at the firm.³⁸ In the same announcement that described the creation of the BDM force, Lanning informed the tax partners that the firm had made a significant investment in telemarketing resources, which would be deployed in connection with products where they could “add significant value.³⁹

Moving on to Greener Pastures

By spring 1997, KPMG had sold ten FLIP deals, earning the firm over $6 million in fees. Another four deals with potential fees of $1.5 million were in the works. FLIP’s original developers, however, were already moving on to other firms. In early 1997, David Lippman, now going by the last name Smith, left KPMG for Quadra, the “investment advisory” firm involved in implementing FLIP. (He was soon joined by Larry Scheinfeld, another tax partner at KPMG, and Ralph Lovejoy, who had been at First Union.)⁴⁰ Schwartz, in the meantime, moved over to the accounting firm Coopers & Lybrand, the predecessor firm to PricewaterhouseCoopers. Smith and Schwartz may have left because they felt that the bureaucratization of the tax product process at KPMG stifled their creativity; perhaps they also had personality conflicts with Pfaff and Larson, who insisted on being present at every sales presentation and were very secretive about the financial aspects of FLIP.⁴¹ Smith and Schwartz were also probably in search of higher income—the management fees paid to investment advisory firms implementing shelters were more than double those earned by KPMG, and the firms were much smaller. Shortly after their departure, Smith and Schwartz began to market a version of FLIP with Coopers & Lybrand and, after the merger with Price Waterhouse, with PricewaterhouseCoopers (PwC), ultimately closing approximately fifty FLIP transactions. Back at KPMG, Larson and Pfaff believed that the strategy was proprietary to the firm and that it had been stolen.

Soon Pfaff and Larson also left KPMG. Although they had no professional investment expertise, they formed their own investment advisory firm under the name “Presidio” whose principal—if not sole—purpose was to collaborate in designing and implementing tax shelters with the tax shelter group at KPMG. Pfaff and Larson anticipated a lucrative long-term relationship between the two firms. Pfaff enjoyed a close friendship with Lanning, dating back to their time together in KPMG’s Denver office. In KPMG, Lanning was known for his confrontational management style, said to reflect his time at the United States Naval Academy. Years earlier, Pfaff became friends with Jeff Stein when they both worked in KPMG’s Paris office, providing tax services to wealthy American expats living in Europe. Stein, a self-confident and abrasive figure—who in the words of one colleague would “sell his mother for the bottom line”—would soon be succeeding Lanning to head KPMG’s tax practice and was putting his full weight behind the firm’s efforts to expand its tax product activities.

To ensure that the process inside the firm progressed smoothly, Pfaff sent Lanning and Stein a lengthy memo setting forth his vision for KPMG’s Tax Advantaged Transaction practice, as the tax shelter group was known, and its relationship going forward with Presidio. According to Pfaff, the goal was to create a “true” turnkey product—a tax solution that worked directly out of the box—on which the firms could earn royalties on sale. To succeed in this endeavor, KPMG needed to forge alliances with the international banking and leasing communities and the small number of law firms that were “skilled and respected in this area.” Pfaff expected that KPMG’s highly respected name would open the necessary doors. Pfaff noted that marketing tax shelters had become more problematic with the U.S. Tax Court decision in ACM because the firm could no longer “openly market” the tax results of an investment.⁴² Pfaff’s takeaway from the opinion was that KPMG had to develop a stealth approach to sales, in which the firm publicly insisted that any tax-advantaged product was an investment opportunity and privately told clients that the real point was tax elimination.

To increase consistency in client engagements and product implementation, Pfaff urged KPMG’s tax leaders to entrust sales and implementation to a handful of tax experts who had the aptitude and interest for tax-advantaged products. “The ideal candidate,” he said, “would be someone who knows the product cold, is flexible, decisive, and has reasonably good sales skills.” Although going with “technicians” was advisable, it was important that “he or she not be a ‘waffler.’” Pfaff identified the KPMG colleagues who had developed experience with FLIP. He also encouraged the tax leaders to rely on a few tax partners—Gregg Ritchie in the L.A. office and Gary Powell in the Denver office—who had proven especially talented at identifying target markets and gaining entry.

Going forward, Pfaff emphasized, he “strongly desire[d] a close relationship with KPMG.” Although a formal alliance would be “subject to scrutiny,” presumably because the two firms would not be able to maintain a facade of independence if the government investigated, Pfaff was eager to enter into an arrangement under which Presidio gave KPMG a right of first refusal on products it had in development in exchange for preferred provider status at KPMG. Pfaff was tempted to recommend that KPMG sever its tie with Quadra in light of the latter’s “open and notorious relationship” with Coopers & Lybrand, but thought the more prudent course was to continue to work with them since Quadra’s cooperation would be needed to complete transactions already in the works and to maintain a united front in the event of an IRS investigation. In any ongoing relationship, however, KPMG had to obtain assurances that its “intellectual capital” was not “pipelined directly to C & L.” Pfaff made clear too that Presidio would be loyal to KPMG in exchange for favored treatment. Special treatment would also be extended to Deutsche Bank, with whom Pfaff had begun to forge relations in the previous months. Pfaff’s hope was to eliminate involvement by UBS and Quadra, the latter of which had brought the former to the table.⁴³ Shortly after Pfaff distributed his memo, the Tax Advantaged Transactions group started to rely on Presidio and Deutsche Bank to implement FLIP. DeLap, and the DPP review committee, which had also received Pfaff’s memo proposing a special relationship, had no objection.⁴⁴

A “Goldman-Sachs”-Type Practice

Upon their departure, Pfaff and Larson encouraged Lanning to put Gregg Ritchie in charge of the FLIP strategy.⁴⁵ Ritchie, a line partner in KPMG’s Los Angeles office, saw great revenue potential among the technology entrepreneurs who were striking it rich in California’s dot-com boom. Ritchie had sold the first few FLIP deals in early 1997 and was enthusiastic about the strategy. Ritchie was also a bulldog who was going to fight to make sure that FLIP and any successor strategies would survive the approval process and get to market. In Ritchie’s own words, he was “more aggressive than most people in the firm about defending my positions and trying to become an advocate for interesting and creative ideas.”⁴⁶ Not surprisingly, he was very successful at sales. Ritchie’s go-getter attitude and his considerable marketing skills had been well rewarded and led to his quick rise in the firm.⁴⁷ Ritchie thought FLIP was “brilliant” and took on the assignment of overseeing its distribution with gusto.⁴⁸

In the fall, Ritchie launched a new group inside the Personal Financial Planning (PFP) practice, Capital Transaction Strategies or “CaTS” whose assignment was to sell FLIP around the country.⁴⁹ He enlisted the participation of various PFP members around the country, including Randy Bickham, an income tax specialist in the Palo Alto office, who soon became an eager collaborator on developing and marketing strategies. Doug Ammerman, the partner in charge of the PFP services had set a $4 million revenue goal for the CaTS group for fiscal year 1998. With FLIP just beginning to generate revenue and delays with obtaining approval for another product in the group’s inventory, Ritchie was concerned that it would be a “significant stretch” to achieve this revenue target.⁵⁰

Ritchie also set about formalizing relations with Presidio and Quadra. He controlled the sales process to ensure that only CaTS members had access to the strategies his group sold. During preliminary planning meanings, he instructed his sales team to present only on a white board and never leave documents with clients. In addition, he insisted that clients should be strongly discouraged from consulting with their legal advisors when they were deciding to purchase FLIP. Ritchie was concerned that the strategy would get into the wrong hands—the hands of someone inside KPMG who would not “understand” the strategy or of someone outside the firm, such as a competitor or perhaps the government.⁵¹

Prior to leaving KPMG in the summer of 1997, Larson had planned to finish all of the outstanding opinion letters on FLIP, but some number remained unfinished. The task of finalizing them fell principally to Bob Simon, a new partner in the International Services group in the Denver office who had joined KPMG from a corporate law firm. Pfaff, who had also been in the Denver office, apparently recruited Simon to the firm, and Simon inherited Pfaff’s reputation for technical know-how.

When Simon first saw the draft opinion for FLIP, he raised questions about the letter itself and about how the strategy was being executed. Many of his concerns focused on whether the transaction had sufficient economic substance. In particular, he was concerned about whether the Cayman entity was more than a “paper company.” He also believed that the taxpayer’s property interest in that entity was not adequate to justify the basis shift and that the foreign bank should bear some economic risk in the transaction.⁵²

Ritchie, for his part, was not especially concerned about the legal weaknesses of the strategy in particular—the fact that its purpose as a tax loss generator was barely disguised. He did not see a meaningful difference between the Dupont-Seagram transaction and FLIP, and wanted to make sure that the strategy had sufficient window dressing so as not to be detected by the IRS on audit.⁵³ As far as Ritchie was concerned, if the firm could conclude that a taxpayer had “50.00001%” likelihood of prevailing in court, that was enough to issue a more-likely-than-not opinion and market the strategy.⁵⁴

While not worried about the legal merits, Ritchie was annoyed at Simon’s meddling. To Ritchie, who was trying to keep sales going at a brisk pace, Simon’s suggestions were obstructionist. To make matters worse, Larry DeLap, the head of the Department of Professional Practice, took Simon’s views seriously.

The conflict came to a head around whether FLIP needed to be registered with the IRS. The 1997 Taxpayer Relief Act had strengthened legislation, dating back to 1986, that required transactions that met certain characteristics to be registered with the IRS so that the agency could examine them to determine whether or not they were abusive transactions. Promoters who failed to register transactions meeting the legislative definition of a shelter were subject to penalties. Simon first raised the issue of whether FLIP had to be registered at a meeting during the fall. Until Simon asked about the registration requirements, no one had thought to look at them. Simon’s question threw the FLIP team for a loop. Ritchie was clear that registering the strategy would not be in the “clients’ best interest.”⁵⁵

As of fall 1997, the Treasury had not issued regulations dealing with the new registration requirements under its authority pursuant to the Taxpayer Relief Act. Simon nevertheless had substantial concerns that FLIP met the definition of a shelter for purposes of the registration requirement that had been in effect since the late 1980s.⁵⁶ After some back and forth, DeLap concluded that KPMG was not required to register FLIP. But in mid-November, he instructed the CaTS team to cease marketing FLIP immediately. Ritchie was none too pleased that the “new kid on the block,” as he called Simon, had persuaded DeLap to halt FLIP.⁵⁷ All told, KPMG sold FLIP to eighty clients, earning approximately $17 million in fees.⁵⁸ FLIP generated $1.9 billion in tax losses.⁵⁹

The CaTS team’s experience with FLIP demonstrated a significant demand for tax elimination strategies. Through their energetic sales efforts, the team had pitched the strategy to a number of clients who were eager to go forward. With the sudden discontinuation of the strategy, Ritchie and his team became concerned that they would not be able to satisfy client interest and would lose out to firms selling similar products. Randy Bickham, an active member of CaTS in the Palo Alto office who worked closely with Ritchie, emphasized:

The existing Foreign Leveraged Investment Strategy product currently has high market demand and we have a backlog of deals. Should we fail to deliver the existing product or a suitable substitute, the CaTs practice will lose the initial market position that has been created with high-wealth individuals to the competition and our commitment to participating in the tax product market will be subject to question by the market and our strategic partners.

If Ritchie and the CaTS team were to aspire to a “Goldman-Sachs”-type practice, Bickham emphasized, “to which high wealth individuals and their advisors [would] immediately look to when liquidating significant portfolio positions,” then the group could not be “out-of-stock” when high-wealth customers came calling.⁶⁰ Earlier in the fall, a team consisting of Ritchie, Simon, Bickham, Ruble of Brown & Wood, and Larson and Pfaff of Presidio had already begun to brainstorm a new strategy. When DeLap pulled FLIP off the shelves, “[t]hese discussions took on an air of urgency,” Stein, who had been following the process closely, later noted.⁶¹

Ritchie and the CaTS members thought a new strategy could be developed by altering some of the features of FLIP. Simon, the technical leader of the development team, was skeptical and argued that a number of FLIP’s weaknesses recurred in the new strategy under development. Simon believed he could develop a strategy that, based on other rules, was less vulnerable to IRS challenge.

As Simon continued to argue for his own approach, Ritchie became increasingly exasperated that the process was dragging. He also questioned Simon’s motivations in bringing up objections and pushing his own strategy. Ritchie suspected that Simon was trying to develop a product that would compete with his. He also believed that Simon wanted a hand in the strategy so that Simon’s group, the International Services practice, would get a share of the revenue credits that would come from marketing the new deal—credits that all should go to Ritchie’s own group, PFP services. In late fall, he decided that the easiest course of action was to cut Simon out of the process altogether. By December, the development group began to participate in conference calls and meetings without including Simon, and soon developed a prototype for the Offshore Portfolio Investment Strategy (OPIS).⁶²

In February 1998, Simon finally got to see OPIS. As he wrote to Ritchie, he thought the strategy failed on both conditions DeLap had set for it: to support a more-likely-than-not level of confidence and be sufficiently different from FLIP that “that we could justify registering the current product (but not FLIP).” Among the nine problems he flagged were that the foreign bank did not incur any economic risk, the deal looked “prewired” from the clients’ perspective, and the Internal Revenue Service would use its anti-abuse authority to disallow the loss. Simply put, OPIS was “Son of FLIP.”

Ritchie would have none of it. As he wrote, “I disagree on your conclusion. We believe we can write a more-likely-than-not and expect DPP to agree.” With regard to Simon’s argument that the IRS was bound to step in, Ritchie remarked: “As we discussed in our conference call, there is simply nothing else to say on this topic. . . . This . . . is one element of why the strategy is only ‘more likely than not.’”⁶³

Ritchie was correct in one regard. Even before OPIS was approved for sale, a dispute had broken out as to how the revenue was going to be credited. Intent on making sure the International Services practice would get its fair share, Stein, the partner in charge, circulated several emails about Simon’s role in developing the strategy. In one email, titled “Simon says,” he listed the many weaknesses Simon had identified in the FLIP strategy. He explained, for example, that Simon was the first to point out that the client had purchased a warrant in the Cayman structure for a “ridiculously high amount of money” and that the warrant “stood out more like a sore thumb since no one in his right mind would pay such an exorbitant price for such a warrant.” According to Stein, “[i]n kicking the tires on FLIP (perhaps too hard for the likes of certain people),”⁶⁴ Simon had discovered that the redemption of shares by the foreign bank from the Cayman company did not occur at the same time as the purchase of the same number of options in the foreign bank by the client. This put the putative tax benefits from the deal in doubt.

Needling Ritchie, Stein pointed out that he had made “an admission against interest” when he earlier had written Simon that OPIS was developed “in response to your and DPP’s concerns over the FLIP strategy.” Stein concluded by emphasizing that development of OPIS had been a team effort. “What I thought we are trying to do here was bringing the best minds we had in this Firm together to design the best product to go to market with.” If the record was examined carefully, Stein claimed, then International Services was responsible for 80 percent of the OPIS product. For the sake of teamwork and collegiality, Stein was willing to stick with 50/50 and “forget about the Ides of March.”⁶⁵

Ritchie was not going to take Stein’s email lying down. “The bottom line,” he wrote to Stein, Ammerman, and other tax leaders, “is, had we allowed Simon to continue to try to develop the OPIS product, we would not have one. Period.” In his opinion, “Jeff’s comments in this message are not in line with the facts, are inflammatory, and I hope are not discoverable by the IRS.”⁶⁶ Ritchie believed that Simon’s proposed strategy had simply lost out. As he insisted, “Larry [DeLap] bounced this effort. Clearly Simon was not creative enough.” Ritchie went on to try to discount every argument Stein made in his memo, concluding, “the facts are that without the PFP practice, the IS practice would still be moaning about not having a product.”⁶⁷ By the time OPIS was approved for sale, the two practices had settled their disagreement and decided to share the revenue credits.⁶⁸

Apart from his annoyance at Simon’s involvement, Ritchie was frustrated that the process of obtaining approval for OPIS was taking months longer than he had anticipated. One wrinkle was that the banking partner on OPIS, brought in through Larson’s contacts, was Deutsche Bank, an audit client of KPMG. Given the rules on auditor independence, DeLap thought it was important to include the Department of Professional Practice Assurance at KPMG to determine how to deal with potential conflicts of interest. KPMG was supposed to be independent of Deutsche Bank, and ethics accounting rules required that the two firms refrain from engaging in certain business relationships. Ritchie was aggravated that DPP Assurance took its time in deciding how to handle potential issues arising out of the relationship with Deutsche Bank and then imposed new rules on the marketing of OPIS. The head of DPP Assurance, for his part, was dismayed with Ritchie’s aggressiveness, about which, he underscored, he had “nothing to say.”⁶⁹

In late spring, the question of whether OPIS had to be registered with the IRS arose. All along DeLap had assumed that this strategy would be registered. As OPIS was finally nearing approval in May 1998, Ritchie sent Jeffrey Zysik, a manager working under him, to investigate how much attention the IRS would give the strategy if it was registered. Zysik came back with good news. He had checked in with Phil Brand and Richard Smith, two partners who had left high-level positions at the IRS the previous year. They, in turn, had contacted IRS officials to find out whether there was a lot of activity following up on shelter registration. The short answer was “no.” Smith had helpfully added that even if the IRS was interested in focusing on registered shelters, it did not have the capacity to match registration forms with partnership returns. There was little downside to registering the shelter.⁷⁰

Despite the signs that the agency would likely not take notice, Ritchie continued strongly to believe that registering OPIS was a bad idea. Putting aside their earlier differences over revenue credits, Ritchie took his case directly to Jeff Stein. Stein, who would soon be promoted to vice chair of Tax Services, would most likely be sympathetic to Ritchie’s point of view and knew how to apply appropriate pressure effectively. In a memo, Ritchie set forth the strongest business reasons he could marshal that the firm should not register OPIS, even if it fell within the registration requirements.⁷¹

As Ritchie explained, his view was based on the “immediate negative impact on the Firm’s strategic initiative to develop a sustainable tax products practice and the long-term implications of establishing . . . a precedent in registering such a product.” More specifically, Ritchie wrote, “the financial exposure to the Firm” from not registering OPIS was “minimal” since any penalties from noncompliance were much smaller than potential profits. At most, they would represent 14 percent of fees earned. Other promoters were not registering their products, so registering the product would put KPMG at a “severe competitive disadvantage” given “industry norms.” In addition, there had been “a lack of enthusiasm on the part of the Service to enforce” the registration provisions. Lastly, Ritchie noted, there was a lack of guidance as to how the registration requirements should be interpreted. All told, he concluded, “the rewards of a successful marketing far exceed the financial exposure to penalties that may arise.”⁷²

When he read Ritchie’s memo, DeLap erupted. In an irate phone call, he explained in no uncertain terms that the firm was not going to decide whether to register OPIS based on economic considerations. Ignoring the registration requirements on this basis could subject individuals or the firm not only to professional sanctions from the IRS but even to criminal liability for engaging in a willful violation of the Treasury’s reporting requirements. One did not decide to disobey rules that reflected professional obligations and imposed penalties for intentional noncompliance based on financial considerations. Citing the relevant statute and regulation, DeLap urged Ritchie to look them up. Even after reading them, Ritchie still did not understand DeLap’s point. Puzzled, he wrote DeLap that he did not see how the registration requirements implicated the ethics rules or criminal code. Since the registration requirements did not make explicit reference to either, how, he wondered, could failing to register subject someone to ethics or criminal sanctions? DeLap, for his part, recommended that OPIS be registered with the IRS, but was overruled by higher-ups in the tax practice.⁷³

In early June 1998, DDP gave its preliminary approval for OPIS to be marketed.⁷⁴ When Ritchie announced the decision to the CaTS team, he reminded members that they were not to leave marketing materials with clients because “it will DESTROY any chance [that] the client may have to avoid the step transaction doctrine.”⁷⁵ As the materials made clear, the purported investment “decisions” were really just prearranged steps with all risk sufficiently hedged and all parties prepared to act as prescribed. On hearing that the strategy was finally approved, Randy Bickham, who had worked closely with Ritchie to obtain approval, was very excited and congratulated him for successfully managing the KPMG organization to get the product to market.⁷⁶ Ritchie, for his part, was less sanguine about the firm’s prospect for becoming a “Goldman-Sachs”-type practice. The firm, he concluded, was too big, too inflexible, and too set in its ways.⁷⁷

In August 1998, Ritchie left KPMG to work as Chief Financial Officer at Pacific Capital Group, a private equity group founded by Gary Winnick. Along with Ken Lay and Bernie Ebbers, Winnick later emerged as one of the poster children for the accounting ploys that that helped drive the bull market in the late 1990s. Winnick created Global Crossing, a company dedicated to creating a global state-of-the-art fiber optic network and a darling of the stock market during the Internet boon. In 2001, the company went bankrupt amid allegations that it had manipulated its financial results, but not before Winnick salvaged $735 million from his investment. Class actions brought by investors against Winnick and his legal advisors settled for a total of several hundred million dollars, and Winnick narrowly avoided being sanctioned by the SEC.

After Ritchie’s departure in August 1998, Jeffrey Eischeid took over direction of the CaTS team. OPIS was marketed through Presidio and Quadra, with Ruble, who had been involved in designing the strategy, providing purportedly “independent” opinion letters. In early December, DeLap sent around an email instructing PFP practice members to discontinue selling OPIS.⁷⁸ A threshold number of sales had been met and DeLap was concerned that if a court discovered that so many clients had done the identical transaction, it would find that the steps of the transaction were prewired and disallow its tax benefits.

Despite DeLap’s instruction that marketing cease, members of the firm continued to market OPIS through 2000. The momentum behind such a lucrative product was impossible to stop. All told, KPMG sold the strategy to at least 170 wealthy individuals, generating approximately $2.3 billion in claimed tax losses and earning more than $28 million from OPIS.⁷⁹

A Shelter for the Little Guy

In 1998, while it was in the middle of launching OPIS, KPMG became involved in another tax shelter known as Short Options Strategy or SOS. The strategy had come to KPMG’s attention via an investment boutique known as Diversified Group Inc. (DGI), which was run by James Haber. At the turn of the century, versions of this strategy were being marketed by Arthur Andersen, Ernst & Young, BDO Seidman, and various law firms and investment advisory boutiques. SOS was one of a family of shelters that later became known as “Son of BOSS” or contingent liability shelters. (Although BOSS, a shelter marketed by PricewaterhouseCoopers, was itself not a contingent liability shelter, and Son of BOSS shelters actually predated BOSS, they acquired this moniker because they came to light after BOSS became public.)

Son of BOSS shelters derived from a shelter that media mogul Ted Turner used to avoid paying taxes when his company went through a reorganization. Although the rules on which Turner relied had subsequently been changed, shelter promoters began to devise strategies based on analogous techniques. Paul Daugerdas was among the first shelter promoters to market these strategies to individuals in 1994 when he was at Arthur Andersen. Emile Pesiri, then a lawyer at Jackson Tufte Cole & Black and a repeat player in the shelter market, provided the opinion letters on Daugerdas’s earliest strategies.

Son of BOSS shelters, which became increasingly popular in the late 1990s, were designed around partnerships or other pass-through entities. Under the Internal Revenue Code, the amount and type of a partnership’s income is calculated using the partnership entity as the relevant unit but the income is taxed to the individual partners. The partners pay taxes on their share of the taxable income of the partnership even if no funds have been distributed to them by the partnership. In the same vein, the partners can deduct a share of their losses incurred by the partnership against any capital gains they earn. These gains do not have to be related to partnership activities.

Generally, the tax code allows a taxpayer to subtract certain costs of acquiring an asset in determining whether a taxpayer has incurred taxable gains on its disposition. These adjusted costs are the asset’s basis. On the one hand, if the sale price of an asset exceeds its basis, the difference between the two is treated as a taxable gain. If, on the other hand, basis exceeds sale price, the taxpayer has a loss that can be deducted against other capital gains. The Son of BOSS shelters all involved transactions designed to increase a taxpayer’s basis in an asset, while avoiding rules that otherwise would operate to reduce the basis. The effect was to generate a loss for tax purposes when the taxpayer sold an asset for less than the amount of the basis—even though the taxpayer suffered no economic loss anywhere near the loss that she claimed for tax purposes.

These shelters created a tax loss without an equivalent economic loss by relying on how different liabilities are treated under the partnership rules. In the case of debt, the partnership rules are straightforward. When a partnership assumes a liability, each partner’s basis—adjusted costs—is increased proportionate to his (or her) share. Suppose that a three-person partnership, in which each member has an equal share, takes out a loan for $30,000. This is treated as if each partner has contributed $10,000 in cash to the partnership. (The transaction is equivalent to each partner borrowing $10,000 and contributing it to the partnership.)⁸⁰ Each partner’s basis therefore goes up $10,000.⁸¹ So if the partnership shows a gain, each partner can subtract the amount of basis from his share of the profit in computing his taxable gain. (Remember, partners are taxed on the gain even if there has been no distribution to the partners.) If the partnership shows no gain, each partner can deduct the amount of his basis from his other capital gains. This approach is consistent with the pass-through treatment of partnerships.

A difficulty arises, however, about how to treat liabilities that are not certain—like debt—but instead are contingent on the occurrence of an event. Consider the following scenario: A partnership has sold an option to purchase land to a bank. Under the option contract, the bank is entitled to buy the property at a specified price through a certain date. The bank, however, is not required to buy the property. Under the tax code, the bank’s payment for the option is considered a decrease in the partnership’s basis, which, under pass-through principles, is ultimately taxable to the partners. If the partnership received $10,000 for the option, each partner’s basis (that is, adjusted costs) would be lowered by a percentage of the $10,000 equivalent to each one’s share of the partnership.

But how should the bank’s right to buy the property under the option contract be treated? Should that right—whose value is reflected in the option payments by the bank—be considered a liability? The Internal Revenue Code’s partnership taxation provisions are silent as to the definition of liability in general and contingent liability in particular. The issue arose in a 1975 case involving a couple named Helmer, who were partners in a cattle business. The Helmers argued that the bank’s right under the option contract should be considered a liability and therefore the price of the option should be reflected in their basis—that is, their adjusted costs—in the partnership. (If the price of the option were included in their basis, they would end up with lower taxable gains.) The Tax Court sided with the IRS, emphasizing that the land options agreement did not require the Helmers to repay any amount if the option was not exercised, nor did it create any other liabilities. There was a possibility that the option would expire and the Helmers would not be obligated to transfer the property to the bank or owe any money.

Through the 1990s, the IRS and courts followed this approach to options sold by partnerships.⁸² Short options, as they are known, created contingent liabilities, which did not increase a partner’s liability interest in a partnership. In contrast, if a partner bought an option—called a long option—and contributed it to the partnership, the cost of the option would be included in the partnership’s basis. Had the Helmers purchased an option to buy land, the payments for the option would have been included in the partnership’s costs and passed through to them individually. The different treatment of long options and short option liabilities in this context makes intuitive sense. On the one hand, when the partnership has bought an option, the partnership is out the payment for the option. Even if the partnership decides not to exercise the option, the price of the option will not be repaid. On the other hand, when the partnership has sold an option, it will incur a cost only if the option is exercised.

The Son of BOSS shelters of the 1990s were designed to exploit this asymmetric treatment of sales and purchases of call options. Consider a simplified example: The taxpayer sells an option to buy securities or foreign currency to a bank (the “short” option). At the same time, the taxpayer buys an offsetting long option from the bank. The price of the long option is slighter higher than that of the short option.⁸³ Because the two transactions offset each other, the taxpayer is not taking an economic position in the transaction. The taxpayer then contributes the long option and the liability incurred under the short option to a partnership and receives a partnership interest in return. After the partnership engages in some pro forma—if highly complex—securities or options trading to give it a facade of economic activity, the taxpayer terminates her interest in the partnership. When the partnership is dissolved, the cost of the contributed long option, now worth nothing, is treated as a loss of the partnership. Under pass-through principles, the taxpayer is entitled to claim this loss against capital gains. (Ideally, the amount offsets most or all of the taxable capital gains the taxpayer would otherwise have to show.)

Meanwhile, the extinguishment of the liability created when the short option expires is ignored for tax purposes. Were this a debt owed by the partnership or some other certain liability, its cancellation would be treated as a decrease of the partnership liabilities and therefore the partner’s share of the liabilities. (In other words, it would be the equivalent of a gain.) But because the obligation under the short option was not treated as a liability when it was transferred to the partnership, the expiration of the option, which fulfilled the obligation, does not decrease the partnership’s liabilities, nor under pass-through principles, a partner’s share of those liabilities. So it would not be subtracted from the loss claimed by the partner—equivalent to the value of the long option going from its purchase price to nothing when the partnership dissolved. The result of this transaction is that the taxpayer can claim losses for tax purposes without having incurred comparable economic losses.

Despite significant contrary authority, including IRS rulings, the partnership anti-abuse rules, and the economic substance doctrine, developers of these shelters took the position that Helmer stood for the proposition that any liability that is contingent—that is, that may not arise or whose amount cannot be calculated at the time it is incurred—should not be included in the calculation of a taxpayer’s basis. In Helmer, tax law precluded using a putative liability to increase a taxpayer’s basis; Son of BOSS promoters claimed that it also precluded using a putative liability to reduce a taxpayer’s basis.

In each instance, of course, the taxpayer is attempting to maximize the amount of her basis. In Helmer, the taxpayers sought to do so by claiming that a noneconomic liability should be counted for tax purposes. In the example involving the short and long options, the taxpayers are attempting to do so by claiming that an economic liability should not be counted for tax purposes. Both efforts represent attempts to claim tax treatment at odds with economic reality. Reading Helmer in light of the Tax Court’s focus on underlying purpose should lead to including the call option obligation in the taxpayer’s basis and then subtracting it from that basis when the option is contributed to the partnership. Promoters of Son of BOSS shelters claimed, however, that the literal terms of the court’s holding in Helmer, not its rationale, should govern.

The SOS shelter KPMG was offering had been developed by R. J. Ruble and Orrin Tilevitz, DGI’s general counsel, who first came up with the idea when he was at Price Waterhouse.⁸⁴ In the DGI version, the taxpayer held the options during a sixty-day period. Because the investment was of such short duration, partners in KPMG’s Washington National Tax office had not been able to conclude that the strategy had a more-likely-than-not probability of being upheld if challenged by the IRS.⁸⁵ Nevertheless, the firm was involved in marketing and implementing the shelter and prepared tax returns for clients who had purchased it.⁸⁶ Because the fees from SOS were comparatively modest compared to those from OPIS, KPMG had been treating SOS as something of a backup shelter. Partners offered the transaction to clients who were not able to meet the capital requirements for its high-end strategies, which required a client who wanted to generate a minimum $20 million tax loss.

When OPIS sales were discontinued, KPMG’s tax shelter team continued to offer SOS. Tax leaders were eager, though, to find a replacement strategy aimed at high-wealth individuals that would prove just as lucrative. A new strategy with potential to generate substantial fees soon emerged.