chevron-down Created with Sketch Beta.

The Tax Lawyer

The Tax Lawyer: Winter 2020

Whose Earnings and Profits? What Dividend? A Discussion Based on the Dr Pepper–Keurig Transaction

Jeffrey T Sheffield

Summary

  • The Article discusses the interplay of the Subchapter C rules regarding earnings and profits (“E&P”), and the timing of dividend payments, with the consolidated return regulations’ rules governing tax year ends, reverse acquisitions, and the replication of E&P mandated by the regulations’ group structure change rules.
  • The payment date for measuring the E&P of the special dividend should be the date of the merger (rather than the actual date of payment) and, contrary to the existing group structure change rules, the relevant E&P should not include any of Keurig’s E&P.
  • The Article asks whether the special dividend should have been characterized as a dividend in the first place
Whose Earnings and Profits? What Dividend? A Discussion Based on the Dr Pepper–Keurig Transaction
Bloomberg Creative via Getty Images

Jump to:

Abstract

In July 2018, Dr Pepper Snapple Group, Inc. (“DPSG”) acquired all the stock of the Maple Parent Holdings Corp. (better known as “Keurig”) in exchange for DPSG common stock. The acquisition was unusual in two respects: DPSG paid its shareholders a pre-merger “special dividend” equal to approximately 87% of the stock’s value; DPSG then issued to Keurig’s shareholders DPSG stock equal to approximately 87% of the combined companies’ stock. This combination of events led to the issues discussed in this Article.

The Article first discusses the interplay of the Subchapter C rules regarding earnings and profits (“E&P”), and the timing of dividend payments, with the consolidated return regulations’ rules governing tax year ends, reverse acquisitions, and the replication of E&P mandated by the regulations’ group structure change rules. From there, the Article argues that the payment date for measuring the E&P of the special dividend should be the date of the merger (rather than the actual date of payment) and, contrary to the existing group structure change rules, the relevant E&P should not include any of Keurig’s E&P.

Second, the Article asks whether the special dividend should have been characterized as a dividend in the first place. Because Keurig was the source of the funds for the special dividend, the Article argues that the reasoning in Waterman Steamship and its progeny could be applied to recharacterize purported dividends as sales proceeds when the funds for the dividends are contributed by outsiders. It goes on to argue that a better approach would be to subject dividends, when coupled with a dilutive stock issuance, to the same type of dividend-equivalence tests developed for determining whether a redemption, or more specifically boot in a reorganization, should be treated as a dividend or sales proceeds for tax purposes. This latter approach would avoid many of the factual issues surrounding application of Waterman Steamship’s source-of-funds approach under current law.

I. Introduction

On July 9, 2018, Dr Pepper Snapple Group, Inc. (“DPSG”), a publicly-traded Delaware corporation and the owner of several iconic consumer brands, acquired all the stock of Maple Parent Holdings Corp. (“Keurig”), a privately-held Delaware corporation that also owned several iconic consumer brands, including the Keurig Coffee brand. Each corporation was the common parent of separate affiliated groups of corporations (the “DPSG Group” and the “Keurig Group”), and each group in all likelihood filed consolidated federal income tax returns. The acquisition was accomplished by having a first-tier, wholly-owned DPSG subsidiary merge with and into Keurig (the “Merger”), with Keurig’s shareholders receiving DPSG common stock (“DPSG Common”) in the transaction and Keurig becoming a wholly-owned DPSG subsidiary.

After the January 29, 2018 signing of the Merger Agreement, the DPSG Common had a trading value of approximately $21.6 billion, and Keurig’s equity had an implied transaction value of approximately $10.6 billion. Based on the relative values and deal structure, one might have expected a transaction in which DPSG acquired Keurig, with Keurig’s shareholders being issued about one-third of DPSG’s fully-diluted DPSG Common.

But this is not what happened, or at least not the entire story. Several days before the Merger, DPSG declared an $18.7 billion cash dividend, equal to roughly 87% of the DPSG Common’s pre-Merger trading value (the “Special Dividend”). The Special Dividend was:

  • Payable to record holders of DPSG Common, determined as of the close of business on the business day immediately preceding the Merger, with the Special Dividend itself made payable (and actually paid) to the pre-Merger DPSG shareholders the day following the Merger;
  • Funded with (1) a $9 billion equity investment that Keurig’s principal shareholder made into Keurig immediately before the Merger and (2) $9.7 billion of loans secured, and drawn down, by Keurig between the signing and the closing of the Merger; and
  • Conditioned, along with the $18.7 billion in financing, on the consummation of the Merger.

With the value of the DPSG Common greatly reduced by the Special Dividend, and the value of Keurig’s equity greatly increased by its principal shareholder’s $9 billion pre-Merger equity investment, the DPSG Common issued to Keurig’s shareholders represented approximately 87% of the DPSG Common outstanding immediately after the Merger. The transaction was described as tax free to the Keurig shareholders under either section 351 or section 368 or both.

The ultimate 87%–13% ownership split suggests that Keurig should have acquired DPSG. Why did the transaction instead take the form that it did, with DPSG acquiring Keurig? The answer appears to be unrelated to any tax issues surrounding the transaction. DPSG’s Common was publicly-traded; Keurig’s was not. Had Keurig acquired DPSG, it would have issued Keurig stock (and cash) to the DPSG shareholders. This stock issuance would have constituted an initial public offering of Keurig stock, necessitating a potentially lengthy SEC registration process. By having DPSG acquire Keurig, the SEC process was accelerated and simplified, and Keurig’s shareholders could nonetheless own a class of publicly-traded stock, providing the Keurig shareholders greater liquidity and flexibility in disposing of their investment after the transaction.

The parties expected the Special Dividend to be characterized as a distribution to DPSG shareholders under section 301, treated as a dividend to the extent of DPSG’s current or accumulated earnings and profits (“E&P”). The parties agreed that the tax status of the Special Dividend as a dividend for tax purposes should be determined by reference to DPSG’s E&P, calculated as of September 30th, the close of Keurig’s taxable year. As so measured, the E&P was a fraction of the total Special Dividend distribution: only about 28% of the Special Dividend was considered a dividend for tax purposes. The remainder of the Special Dividend was considered, pursuant to section 301(c), as first a reduction in a shareholder’s tax basis in her DPSG stock, and then as a sale or exchange of her stock (determined on a shareholder-by-shareholder basis).

This Article focuses on the tax treatment of the Special Dividend. The Article initially assumes that the Special Dividend is correctly characterized as a section 301 distribution for federal income tax purposes and asks which corporation(s)’ E&P, and over what period of time, is or should be relevant for the purposes of determining the tax status of the Special Dividend. The Article then takes up the broader question of whether the Special Dividend should in fact be characterized as a section 301 distribution or instead be treated as a sale of stock by DPSG’s historic shareholders.

The DPSG-Keurig transaction was clearly a bona fide business transaction motivated largely, if not entirely, by nontax considerations. The documents reflect the careful work of intelligent, seasoned transactional lawyers. The specific form of the transaction—while unusual—could be easily justified on nontax grounds: by ensuring that DPSG was the surviving corporation in the transaction, the two shareholder groups each received (or retained) stock in a publicly-traded company, a result not as easily or quickly obtained through any other structure. The parties in the DPSG-Keurig transaction adopted thoughtful positions regarding the tax treatment of the Special Dividend necessitated by the transaction structure adopted. Characterizing the Special Dividend as a section 301 distribution did not appear to deliver significant tax advantages to any particular DPSG shareholder group.

The concerns developed herein are not that the parties acted inappropriately, but that the law needs both clarification and correction. This Article argues that if the Special Dividend is respected as a dividend distribution for tax purposes then: (1) the E&P pool for determining the Special Dividend’s tax status should be calculated using DPSG’s taxable year, which under the consolidated return regulations ended on the July 9th Merger date; and (2) DPSG’s E&P should not include any of Keurig’s E&P for purposes of determining the tax status of the Special Dividend. This Article argues further that the Special Dividend should not be characterized as a dividend distribution for tax purposes in the first place, but should instead be characterized as a sale of DPSG Common pursuant to tests for nondividend equivalence already applied to other types of corporate transactions.

II. Corporate Distributions and E&P

This Part briefly reviews the current tax treatment of corporate distributions and the related concepts of current and accumulated E&P.

A. Corporate Distributions

1. Tax Treatment of Corporate Distributions

A corporation’s distribution of property to its shareholders with respect to their stock is, under section 301(a) and (c), considered a dividend to the shareholders, to the extent paid out of the distributing corporation’s E&P.

Dividend income received by a noncorporate domestic shareholder generally is taxed at long-term capital gains rates of up to 20%, provided the shareholder has held her stock for a specified period (generally more than 60 days out of a 121-day period that begins 60 days before the record date of the dividend distribution). If the shareholder does not satisfy the holding period requirement, the dividend is taxed as ordinary income at rates up to 37%. If the dividend constitutes a large “extraordinary” dividend eligible for a reduced tax rate (which the Special Dividend would likely constitute), then a later loss on the sale of the stock must be recharacterized as a long-term capital loss, up to the amount of the extraordinary dividend.

Dividend income received by a domestic corporate shareholder will be taxed as ordinary income, at rates of up to 21%. If, however, the recipient corporate shareholder has held its stock for a specified period (generally more than 45 days out of a 91-day period that begins 45 days before the record date of the dividend distribution), then the dividend is eligible for a 50% dividends-received deduction, reducing the effective tax rate to 10.5%. If the dividend constitutes a large “extraordinary” dividend eligible for a dividends-received deduction (which the Special Dividend likely constituted) and the shareholder has not held the stock for more than two years before the dividend declaration date, the shareholder’s basis will be reduced by the nontaxed portion of the dividend, with any excess treated as gain from a sale or exchange of the stock (likely as capital gain).

Dividend income received by a nonresident alien or foreign corporation (not engaged in a United States trade or business) will be taxed at a 30% rate, subject to reduction under any applicable tax treaty with the United States.

To the extent the shareholder receives a distribution in excess of the shareholder’s allocable share of the distributing corporation’s E&P, the distribution first reduces tax basis in shares and then constitutes gain from a sale or exchange of stock. The gain will be short-term or long-term, depending on the shareholder’s holding period for the stock on the payment date for the distribution. Domestic individual shareholders are taxed at a preferential 20% rate to the extent the gain qualifies as long-term capital gain. Domestic corporate shareholders are taxed at the same 21% rate on both short-term and long-term capital gains. Foreign shareholders will generally not be subject to United States taxation on their capital gain, whether short-term or long-term.

As the above demonstrates, the tax consequences of characterizing a distribution as a dividend—as opposed to, say, sale proceeds or taxable property received in a reorganization—are neither uniformly positive nor uniformly negative for taxpayers. Much depends on whether the recipient is an individual or corporation (domestic or foreign) and, in the case of domestic holders, whether their holding period is sufficient to capture either the reduced rate for dividends or the dividends-received deduction and, to the extent of gain recognition by individuals, whether the holding period is sufficient to qualify for long-term capital gains rates. DPSG did not appear to have any large, dominant shareholders who would have been particularly advantaged (or disadvantaged) by having the Special Dividend treated as a section 301 distribution for tax purposes or by maximizing (or minimizing) the amount of E&P relevant to the distribution.

2. Measuring E&P

The Code does not define E&P. Section 312 itself contains only rules relating to adjustments to E&P, not the original calculations thereof. The closest attempt at defining E&P is found in the regulations to section 312, but even there the definition is reasonably vague (“due consideration must be given to the facts . . . “); and because the regulations explain section 312, they naturally focus more on adjustments to E&P as opposed to original calculations thereof. This Article will not delve into the rules for calculating a corporation’s E&P, but will simply assume that any given corporation has correctly calculated its stand-alone E&P.

The consolidated return regulations contain additional rules regarding E&P. Under those rules, each group member computes its stand-alone E&P. Stand-alone E&P is then “tiered up,” from lower-tier to upper-tier subsidiaries, until ultimately the common parent’s E&P consists of its stand-alone E&P plus the aggregate (net) E&P of all its lower-tier subsidiary group members earned while members of the group. In effect, the entire group is treated as a single entity for E&P purposes, with the common parent’s E&P generally equal to the E&P it would have had were each of its subsidiaries operated as a branch of the common parent.

A corporation’s E&P includes both E&P accumulated in years prior to the taxable year of the distribution and current E&P earned separately in the taxable year of the distribution. Current E&P is “computed as of the close of the taxable year without diminution by reason of any distributions made during the taxable year . . . [and] without regard to the amount of the earnings and profits at the time the distribution was made.” Dividends will be considered first out of current E&P, to the extent available, and then out of accumulated E&P. As a consequence, a dividend declared during a corporation’s taxable year may be considered out of E&P even if, for example, the distributing corporation has a deficit accumulated E&P account, but positive current E&P. More significantly for our purposes, a mid-year corporate distribution can be treated as one “out of” E&P based on E&P generated after the distribution is made (but before the close of the distributing corporation’s taxable year).

B. Determining the Tax Status of a Corporate Distribution

As the above discussion indicates, the tax status of a corporate distribution is based on the results of three important, interrelated inquiries:

  1. What is the end of the distributing corporation’s taxable year?
  2. What is the distributing corporation’s E&P at the close of that taxable year?
  3. In which taxable year is the distribution made?

Part III of this Article addresses how to establish taxable years and E&P when two consolidated groups of corporations combine, as occurred in the DPSG-Keurig transaction. Part IV addresses the surprisingly knotty issue of determining the taxable year in which the distribution is made for purposes of measuring the distributing corporation’s E&P.

III. Taxable Years and E&P When Two Consolidated Groups Combine

When two consolidated groups of corporations combine, the effects on taxable year and E&P vary significantly, depending on whether the combination is characterized as a “reverse acquisition” under the consolidated return regulations. As discussed below in more detail, a reverse acquisition occurs when the stock (or assets) of the common parent of one consolidated group is acquired by another consolidated group, but the shareholders of the first group’s common parent receive stock of the acquiring group’s common parent that is more than 50% of the acquiring common parent’s stock.

A. No Reverse Acquisition

In the examples below, X and Y are each the common parent of an affiliated group of corporations filing consolidated federal income tax returns with December 31 taxable years (the “X Group” and “Y Group”), and the shareholders of X and Y are unrelated to each other. The reverse acquisition rules do not apply in these first two examples because Y’s shareholders do not receive a majority of X’s stock in exchange for their Y stock.

Example 1: On March 31, X acquires all of Y’s stock. Y’s shareholders receive solely X stock constituting 25% of X’s outstanding post-transaction stock. Immediately before the acquisition, X’s E&P is $100, and Y’s E&P is $400.

Y and its subsidiaries have entered the X Group, because all their stock is now owned, directly or indirectly, by X. Under Regulation section 1.1502-76, the taxable year for Y (and all other members of the Y Group) terminates when, as a result of the merger, they enter the X Group and thus cease to be a member of the Y Group. More specifically, the taxable year for each member of the Y Group “ends for all Federal income tax purposes at the end of [the day it joins a new group or ceases to be a member of the old group].” So in Example 1, the taxable year of Y and each Y Group subsidiary ends on the close of business, March 31.

However, consistent with the result when one corporation acquires the stock of another corporation in a transaction not governed by the consolidated return regulations, the E&P of both the Y Group and the X Group remains unaffected by the transaction. X’s E&P remains at $100, and Y’s E&P remains at $400. The same results apply whether X’s acquisition is taxable or tax free to Y’s shareholders.

Example 2: On March 31, Y merges into X in a reorganization under section 368(a)(1)(A) (an “A reorganization”), with X surviving the merger and acquiring substantially all of Y’s assets. Y’s shareholders receive solely X stock constituting 25% of X’s outstanding post-merger stock. Immediately before the acquisition, X’s E&P is $100, and Y’s E&P is $400.

Once again, Y and each member of the Y Group enters the X Group. The taxable year of each Y Group member ends on the close of business on March 31, the date Y merges into X. The result as to Y is consistent with the Subchapter C rules relevant when one stand-alone corporation merges into another corporation in a nonrecognition transaction described in section 381(a)(2). Pursuant to these Subchapter C rules, the transferor corporation’s tax year “shall end with the close of the date of . . . transfer.”

In Example 2, section 381, rather than the consolidated return regulations, controls the treatment of X’s and Y’s E&P. And contrary to the results in Example 1, since the transferor corporation (Y) merges into the acquiring corporation (X) in a section 381 transaction (here, an A reorganization), the acquiring corporation inherits the acquired corporation’s E&P pursuant to section 381(c)(2). So in Example 2, X’s E&P is increased from $100 to $500 as a result of the merger. Y’s E&P is added to X’s accumulated E&P at the close of business on the merger date (i.e., when Y’s taxable year ends as a result of the merger). Adding the E&P at the close of Y’s taxable year permits adjustments to Y’s E&P for activities undertaken by Y on the merger date itself. For example, were Y to pay a dividend to its shareholders on the merger date, but before the merger, the amount of E&P inherited by X would be calculated after taking into account this merger-date dividend.

B. Reverse Acquisition

1. Reverse Acquisition Defined

The consolidated return regulations define a reverse acquisition using somewhat convoluted language. Its central teaching, however, is reasonably clear: when two previously-separate consolidated groups combine, the group whose shareholders own a majority of the stock of the common parent of the combined group continues, and the other, “smaller” group terminates. This is true even if, contrary to relative size, it is the common parent of the smaller group that in form acquires the common parent of the larger group.

Example 3: On March 31, X acquires all of Y’s stock. Y’s shareholders receive solely X stock constituting 60% of X’s outstanding post-transaction stock. (Alternatively, on March 31, Y merges into X in an A reorganization, with Y’s shareholders receiving 60% of X’s stock.) Immediately before the acquisition, X’s E&P is $100, and Y’s E&P is $400.

Whether structured as a stock acquisition or an asset acquisition, each form of corporate combination described in Example 3 constitutes a reverse acquisition of the X Group by the Y Group. As will be seen below, the reverse acquisition rules have an important effect on the tax positions of the affected groups, especially regarding taxable years and E&P.

2. Effect of Reverse Acquisition Rules on Taxable Years

Absent Regulation section 1.1502-75(d)(3), in either version of Example 3 above the Y Group would terminate (it enters the X Group), and the X Group would continue. Significantly, each version of Example 3 satisfies the definition of a reverse acquisition: Y’s stock (or substantially all of Y’s assets) is acquired by the X Group, and the former Y shareholders receive more than 50% of X’s stock in the transaction. Because the transaction is a reverse acquisition, the X Group is deemed to have entered the Y Group—even though in form Y and its subsidiaries entered the X Group as new X subsidiaries—and it is the X Group that terminates. By treating the X Group as the terminating group: (1) the taxable year of all X Group members ends at the close of business on the date X acquires the Y Group; (2) the taxable year of the Y Group members continues; with (3) X “taking the place” of Y as the common parent of the Y Group.

So, in Example 3, each X Group member’s taxable year ends at the close of business on March 31, and the X Group files its final consolidated return for the period January 1 through March 31. The historical Y Group subsidiaries’ taxable year continues through December 31. When Y does not cease to exist in the reverse acquisition (e.g., X acquires Y’s stock), Y continues to report its income and loss for the period January 1 through December 31 in the Y Group’s return. The new common parent X reports its post-transaction income and loss for the period April 1 through December 31 in the Y Group’s calendar-year return.

3. Effects of Reverse Acquisition Rules on Common Parent’s E&P– Group Structure Change

When, as in Example 3, the transaction constitutes a reverse acquisition, the transaction will also be considered a “group structure change.” Under the group structure change rules, the E&P of the common parent of the continuing group is added to the E&P of the new common parent of the group.

Requiring that the new common parent include the old common parent’s E&P was thought necessary to prevent the new common parent from making potentially tax-free dividend distributions following a reverse acquisition. In the paradigm situation, a newly-formed shell corporation with no tax history becomes the new common parent for the group by acquiring all the stock of the old common parent in exchange for 100% of the new common parent’s stock. Without the group structure change rules, the new common parent would have no E&P and could make distributions to its shareholders (all of whom, in the paradigm case, would be old common parent shareholders) unburdened by the old common parent’s E&P.

Under the group structure change rules, the two corporations’ E&P is combined pursuant to Regulation section 1.1502-33(f)(1)(i), a provision worth quoting at length:

[T]he earnings and profits of [X] are adjusted immediately after [X] becomes the new common parent to reflect the earnings and profits of the former common parent [Y] immediately before the former common parent [Y] ceases to be the common parent. The adjustment is made as if [X] succeeds to the earnings and profits of the former common parent [Y] in a transaction described in section 381(a).

So, in Example 3, X’s E&P increases from $100 to $500 immediately after X acquires Y’s stock. As will be discussed in Part V.C below, the requirement that the new common parent succeed to the old common parent’s E&P immediately after the acquisition has important implications to the calculation of DPSG’s E&P.

IV. The Date for Measuring the E&P of a Corporate Distribution

When a corporate board declares a dividend (the “declaration date”), it normally establishes the following: (1) the amount of the dividend; (2) the date designating the “shareholders of record” who are entitled to receive the dividend (the “record date”); and (3) the date for payment of the dividend (the “date of payment”). In general, although the law can vary from state to state, the declaration of a dividend creates a debtor-creditor relationship between the corporation and the shareholder until payment is made. But like any other contract, the dividend declaration may be subject to certain conditions. Dividends can generally only be declared out of legal surplus. Dividends can also be subject to fact-specific conditions, as occurred in the case of the Special Dividend. But the emphasis in the case law is to find an obligation to make a dividend payment at some point before the actual payment of the dividend.

The tax question of when a corporate distribution is considered to have been made quickly becomes acute when the distribution occurs just before—or just after—the close of the corporation’s taxable year. In the classic example, a calendar-year corporate taxpayer declares a dividend on December 27, Year 1, “payable” to its shareholders of record on December 30, Year 1, but paid by check deposited in the mail late on December 31 and received by calendar-year shareholders in January, Year 2. Which year(s) are relevant for purposes of determining the extent to which the distribution is “out of” the corporation’s E&P?

It should initially be understood that a shareholder in the above example must include the dividend in her taxable income in Year 2, when the check is received, not in Year 1 when the check is mailed. This is because dividend income is recognized using the cash basis of accounting, regardless of the recipient’s general method of accounting. Since the shareholders had no access to the check other than by mail, and since the mail would not be delivered until Year 2, the later year is the appropriate year in which to tax the shareholders on their dividend income.

But the shareholders’ tax treatment does not control when the corporation is considered to have made the distribution for purposes of testing the tax status of the distribution. The Service first addressed this latter issue in Revenue Ruling 62-131, which held that:

The date of payment, rather than the date of declaration, constitutes the date of distribution of a dividend. Accordingly, the taxable status of a distribution and its effect on the earnings and profits of the declaring corporation will be determined by reference to the earnings and profits of the corporation for the corporation’s taxable year of payment.

So when is the “date of payment”? The Service addressed this issue in Revenue Ruling 65-23, in which a calendar-year corporation declared a dividend on December 2, 1963, to shareholders of record on December 13, 1963. The corporation paid the dividend by issuing checks mailed late in the afternoon of December 31, 1963, so that no shareholder could receive a check until January 1964. The Service concluded that:

The taxable status and the effect on earnings and profits of the paying corporation of distributions [under the circumstances described above] is determined by reference to the earnings and profits of the taxable year ending December 31, 1963, or the accumulated earnings and profits at December 31, 1963, of the paying corporation.

The Revenue Ruling contains no further explanation, but appears to be based on the fact that the corporation parted with the funds no later than December 31 when it mailed the dividend checks. By December 31, the corporation had taken all steps necessary to pay the dividend. All that remained was the ministerial step of delivering the dividend checks to shareholders. While shareholder taxation of dividends may be based on the cash method of accounting, the corporation’s date of payment for E&P purposes seems to be based more on accrual method of accounting principles. With all steps completed in 1963, there was no reason for the corporation’s earnings in 1964 to be included in the E&P calculation for the dividend.

V. Determining the Tax Status of the Special Dividend

With the groundwork now laid, we turn to the first issue presented by the Special Dividend paid by DPSG to its pre-Merger shareholders: how do we calculate the tax status of the Special Dividend?

A. Effect of the Merger on Taxable Years and E&P

1. Effect of the Merger on DPSG’s and Keurig’s Taxable Years

The Merger clearly qualifies as a reverse acquisition: Keurig’s shareholders received more than half (specifically, 87%) of DPSG’s stock in exchange for their Keurig stock. Under the reverse acquisition rules, the DPSG Group is the terminating group, and the Keurig Group is the continuing group. More specifically, as a result of the Merger:

  • The DPSG Group terminates. The taxable year of DPSG (and every other member of the DPSG Group) ends on the close of business on July 9th, the date of the Merger. This group would file its final consolidated return covering the short period from January 1st through the close of business on July 9th;
  • The Keurig Group continues in existence. The taxable year of Keurig (and every other member of the historic Keurig Group) is unaffected by the Merger. This group’s 2018 consolidated return would include the operations of the historical Keurig Group from January 1st through September 30th, and the operations of the former DPSG Group from July 10th through September 30, 2018; and
  • DPSG is the new common parent of the continuing Keurig Group.

2. Effect of the Merger on DPSG’s and Keurig’s E&P

Since the Merger qualifies as a reverse acquisition in which the former common parent continues in existence, it also qualifies as a group structure change. Under these rules, as discussed in Part III above, DPSG’s E&P is “adjusted immediately after” DPSG becomes the new common parent to include Keurig’s E&P. This appears to be the case even though DPSG’s taxable year has not yet ended, and DPSG does not enter the Keurig Group “for all Federal income tax purposes” until the following day, July 10th.

As a result, DPSG seems to have E&P equal to the combined DPSG-Keurig E&P on July 9th, immediately after the Merger but before the close of business on July 9th.

B. What Is the “Date of Payment” for the Special Dividend?

Under Revenue Ruling 62-131, “[t]he date of payment, rather than the date of declaration, constitutes the date of distribution of a dividend” for purposes of measuring the distributing corporation’s E&P. The tax status of the Special Dividend thus depends critically on whether its date of payment is July 9th or July 10th: if July 9th, then none of the post-Merger E&P of the Keurig-DPSG Group would be relevant; if July 10th, then the relevant E&P would be measured at September 30, 2018, the end of the Keurig Group’s (continuing) taxable year.

1. DPSG’s Position

The proxy statement includes the following disclosure about the tax status of the Special Dividend:

the special cash dividend will be considered a dividend for U.S. federal income tax purposes to the extent of the current and accumulated earnings and profits of DPSG and [Keurig] through the end of the taxable year of [Keurig] in which the merger occurs (which is assumed to be the year ending September 30, 2018) (“E&P”).

Based on this disclosure, the parties apparently decided that, under Revenue Ruling 65-23, the taxable status of the Special Dividend should be determined by reference to DPSG’s E&P, measured at the end of Keurig’s September 30, 2018 tax year. In order to reach this result, they must have determined that, for purposes of measuring DPSG’s E&P, the Special Dividend’s date of payment was July 10th, the day after the Merger. For if the Special Dividend were considered payable before that date, it would have been measured by reference to DPSG’s taxable year ending July 9th, the day DPSG’s taxable year ended when it entered the Keurig Group under the reverse acquisition rules.

2. Detailing the Timing of the Special Dividend Payment

To help decide whether July 9th or July 10th should be considered the Special Dividend’s date of payment, it will be helpful to present the following details on the Special Dividend’s payment mechanics:

  • On June 26, 2018, DPSG’s board of directors (the “DPSG Board”) conditionally declared the Special Dividend and established July 6, 2018 (a Friday) as the record date for the Special Dividend. The Merger date was anticipated to be on July 9, 2018. The Special Dividend was to be made payable to the record holders at 12:01 a.m. EDT on July 10, 2018, one business day after the closing of the Merger, to the July 6, 2018 record holders.

    Payment of the Special Dividend was specifically conditioned upon the closing of the transaction. This was necessary because, as discussed more fully in Part VI.A below, DPSG did not have the funds necessary to pay the Special Dividend absent the Merger (nor would DPSG have wished to pay such a large dividend absent completion of the Merger).
  • On June 29, 2018, DPSG’s shareholders approved the transaction. The shareholders did not vote to approve the Merger directly. Rather, they voted to approve the issuance of DPSG Common as Merger consideration and to increase the number of authorized shares of DPSG Common (which increase was necessary to permit Keurig shareholders to receive the number of DPSG shares they were entitled to receive under the Merger Agreement).
  • On July 9, 2018 (a Monday), DPSG’s acquisition subsidiary merged into Keurig. At the moment the Merger became effective with the Delaware Secretary of State (the “Effective Time”), each share of acquisition subsidiary stock was converted into a share of Keurig stock, and each share of Keurig stock was converted into a fixed number of shares of DPSG Common. Immediately after the Effective Time, DPSG owned 100% of Keurig’s stock, and the former Keurig shareholders owned approximately 87% of the outstanding DPSG Common. Of course, the DPSG Common stock received by Keurig shareholders was “ex-dividend”; the Keurig shareholders were not entitled to receive any part of the Special Dividend.

    Also, at the Effective Time, pursuant to the Merger Agreement, the DPSG Board was reconstituted so as to consist of two previously-serving members of the DPSG Board (the “Continuing Members”), eight members appointed by Keurig, and two independent directors.
  • Still on July 9, 2018, immediately after the Effective Time, DPSG deposited with Computershare Trust Company, N.A. (“Computershare”), its previously-designated third-party paying agent, the funds necessary to pay the Special Dividend. By the time the funds were contributed, all conditions precedent to the payment of the Special Dividend had been either waived or satisfied.

    While DPSG’s agreement with Computershare does not appear to be publicly available, these are generally straightforward agreements under which, in cases such as this, DPSG would appoint Computershare as its agent to make a dividend payment. The distributing corporation must supply the paying agent with the funds needed to make the dividend payment, normally at least one day before payment occurs.

    The Merger Agreement effectively prevented the newly-installed DPSG Board from rescinding the Special Dividend. Section 7.16 of the Merger Agreement required the affirmative vote of both the two Continuing Directors for the DPSG Board to:
[A]mend, modify, or waive . . . [any of the provisions regarding the Special Dividend] or take any action in connection with the Special Dividend. . . . The Continuing Directors . . . shall have the authority . . . to institute any action on behalf of DPSG or the Dividend Stockholders [i.e., the pre-Merger DPSG stockholders] to enforce obligations arising under or in connection with this Agreement.
  • At 12:01 a.m. EDT on July 10, 2018, the Special Dividend became payable to DPSG stockholders who were stockholders of record on the July 6, 2018 record date. The paying agent then distributed the cash previously deposited with it by DPSG to the record holders entitled to receive the Special Dividend.

Why not make the Special Dividend payable on July 9th, immediately after the Effective Time? It is doubtless that payment on that date would have occurred if the Special Dividend had instead taken the form of sales proceeds or boot, since payment would have been included as part of the merger consideration. Payment immediately after the Effective Time also would have minimized the need to protect against possible rescission of the Special Dividend by the (post-Merger) DPSG Board, however unlikely. The reason for the delay may lie with the NYSE rules regarding ex-dividend trading. Under these rules, ex-dividend trading for extraordinary dividends begins on the payment date for the dividend. If the Special Dividend’s payment date were set for July 9th (the anticipated Merger date), the DPSG Common apparently would then have begun trading ex-dividend at the opening bell on July 9th. But this might well have been before the Effective Time of the Merger. And if for some reason the Merger were to be delayed (or possibly even abandoned), the ex-dividend trading would have generated significant problems, to say the least. Setting the Special Dividend’s payment date for 12:01 a.m. the day after the Merger date neatly avoided this potential problem, but gave rise to the tax issue discussed immediately below.

3. Determining the Payment Date of the Special Dividend for E&P Purposes

What was the “date of payment” of the Special Dividend for purposes of calculating E&P? The facts rather clearly support July 9th. DPSG was irrevocably committed to paying the Special Dividend once the Merger occurred on that date. At the Merger’s Effective Time, DPSG obtained the funds necessary to pay the Special Dividend, and those funds were immediately transferred to the paying agent. At that point, it was simply not possible to rescind the obligation to pay the Special Dividend. To prevent any backsliding, the Merger Agreement included specific contractual provisions designed to ensure that the Special Dividend was in fact paid. Failure to pay the Special Dividend—and retention by DPSG of the $18.7 billion in funds earmarked for its payment—would in any event have created a massive windfall to the former Keurig shareholders. In that situation, this group would then own 87% of the stock of a corporation that still held the $18.7 billion that was supposed to be paid entirely to the pre-Merger DPSG shareholders. No court would countenance such a brazen change of heart.

In this context, Computershare was simply a contemporary stand-in for the U.S. Postal Service, a messenger for the delivery of the Special Dividend. The teaching of Revenue Ruling 65-23 is that the E&P should be computed as of July 9th, when DPSG irrevocably transferred the funds to Computershare, and the only action left was for Computershare to deliver the funds to DPSG’s historic shareholders. But July 9th was the last day of DPSG’s first-half 2018 taxable year, with any E&P generated after that date irrelevant in determining the tax status of the Special Dividend. The tax status of the Special Dividend thus should have been determined as of July 9th, even though the Special Dividend was not paid to shareholders until 12:01 a.m. on July 10th.

C. Should DPSG’s E&P Pool for the Special Dividend Include Keurig’s Pre-Merger E&P?

Under the group structure change rules, as quoted above, DPSG’s E&P was “adjusted immediately after” DPSG became the new common parent on July 9th to include the E&P of Keurig on that date, the former common parent of the continuing Keurig Group. This appears to be the case even though DPSG’s taxable year does not end until the close of business on July 9th, and DPSG did not enter the Keurig Group until July 10th.

As a result of the “adjusted immediately after” language, it seems clear under the group structure change rules that DPSG’s E&P equaled the combined DPSG-Keurig E&P on July 9th, determined immediately after the Effective Time of the Merger. As discussed above, if DPSG were a newly-formed shell corporation, this may well be the proper result: had the former Keurig shareholders received a pre-Merger distribution from the old common parent (Keurig), the distribution would be considered a dividend out of Keurig’s E&P. On July 9th, former Keurig shareholders owned the bulk of the new common parent’s (DPSG’s) stock. Interposing DPSG between Keurig and its shareholders should not change the result as it pertains to the former Keurig shareholders. The group structure change rules are designed to ensure that the tax status of distributions received by them from the new common parent include the old common parent’s E&P.

But in this situation, the Special Dividend was received by none of Keurig’s former shareholders—that is, by none of the shareholders that the group structure change rules were designed to capture. Rather, only the pre-Merger DPSG shareholders, who had no ownership in Keurig (or claim to its E&P) before the Merger and who own a distinct minority interest (13%) in the combined company after the Merger, received the Special Dividend. From the perspective of DPSG’s historic shareholders, this is not a change in the structure of the DPSG Group but is instead an acquisition of DPSG by an unrelated corporation. That is certainly the teaching of the reverse acquisition rules. In this case, it makes little sense for a different part of the consolidated return rules to require that Keurig’s E&P be taken into account in determining the tax status of the Special Dividend paid to historic DPSG shareholders.

This disconnect arises from the inconsistent approaches taken by the reverse acquisition rules and the group structure change rules. These rules (1) treat the Keurig Group as the continuing group, which would suggest that the transaction should, for purposes of combining E&P, be tested as if DPSG merged into Keurig; but then (2) combine the two groups’ E&Ps as if Keurig merged into DPSG in a section 381 transaction. Had the corporations’ E&P been combined by assuming—consistent with the theory of the reverse acquisition rules—that DPSG merged into Keurig in a section 381 transaction then (1) DPSG’s year would have ended on the Merger date, (2) the Special Dividend would presumably have been paid on or before the Merger date (since DPSG’s shares would have gone out of existence when it merged into Keurig), and (3) DPSG’s E&P for purposes of testing the Special Dividend would not have included any of Keurig’s E&P. Is this not the more sensible answer?

DPSG’s position regarding the E&P pool for the Special Dividend—that the E&P pool should include the E&P of the combined DPSG-Keurig Parent group, determined as of the end of September 30, 2018, is certainly defensible, and may be the only possible reading of existing group structure change rules for combining E&Ps. But as argued above, the position that the Special Dividend was payable the day following the Merger is too formalistic, and the rule that includes any of Keurig’s E&P in the calculation of DPSG’s E&P illustrates a likely defect in the consolidated return regulations. This is a classic case of a well-intentioned rule producing unintended consequences when applied to a different set of facts than those assumed when the rule was first developed.

Stepping back from the situation, the overall economics of the Merger are that the historic DPSG shareholders’ percentage ownership in DPSG is reduced from 100% to 13% as a result of the Special Dividend, Merger, and related transactions. This fact opens up the discussion of the more general question addressed in the next Part: for tax purposes, was the Special Dividend really a dividend distribution at all?

VI. Was the Special Dividend a Dividend for Tax Purposes?

DPSG’s proxy statement wobbles a little when describing the tax status of the Special Dividend:

For U.S. federal income tax purposes, the special cash dividend is expected to be characterized as a distribution pursuant to Section 301(a) of the Code. Assuming this characterization applies, as a result, the special cash dividend will be considered a dividend for U.S. federal income tax purposes to the extent of the current and accumulated earnings and profits of DPSG and [Keurig] through the end of the taxable year of [Keurig] in which the merger occurs. . . . No tax opinion will be provided to DPSG or the holders of DPSG common stock in connection with the Transactions.

This statement hesitates around classifying the Special Dividend as a dividend distribution for tax purposes. As discussed below, while there is support under current law for treating the Special Dividend as a dividend distribution for tax purposes, there is also good reason to hesitate.

A. Detailing the Financing of the Special Dividend Payment

Part V.B.2 detailed the timing of the Special Dividend payment but deferred further discussion of the financing of that payment. We now take up the financing: how exactly did DPSG obtain the $18.7 billion needed to pay the Special Dividend? The proxy statement and related documents disclose the following:

  • On January 29, 2018, the date the Merger Agreement was signed, Keurig obtained an equity commitment letter from its principal shareholder, under which the shareholder agreed to purchase, immediately prior to the Merger, additional Keurig equity aggregating $9 billion.
  • On February 28, 2018, Keurig entered into a $2.7 billion term loan agreement and a $2.4 billion revolving credit agreement. Upon consummation of the Merger, the obligations under these agreements were contractually assumed by DPSG, with DPSG and all of its subsidiaries (including still Keurig and its subsidiaries) generally becoming liable as borrowers or guarantors under the term and revolving credit agreements.
  • On May 25, 2018, Keurig (through an escrow issuer) issued a series of senior notes due between 2012 and 2048, aggregating $7.75 billion. Upon consummation of the Merger, the obligations under the senior notes were contractually assumed by DPSG, with DPSG and all of its subsidiaries (including still Keurig and its subsidiaries) generally becoming liable as borrowers or guarantors under the senior notes.
  • On July 9, 2018, Keurig transferred $18.7 billion to DPSG. As discussed above, DPSG then deposited the $18.7 billion with Computershare to fund the Special Dividend payment.

One fact may be immediately apparent: the funds obtained to pay the Special Dividend came entirely from Keurig. In fact, Keurig was the party legally required to obtain the necessary financing. DPSG’s obligation to close the Merger was explicitly conditioned on:

[Keurig] having obtained the financing related to the merger at or prior to the closing date of the merger and the funding of such financing in accordance with the terms and conditions thereof at or prior to the closing of the merger, which proceeds, together with other immediately available and unconditional funds, will be sufficient to fund the transactions contemplated by the merger agreement and related fees and expenses.

Keurig’s failure to provide the necessary financing would trigger the payment of a $700 million termination fee by Keurig to DPSG.

There is also no doubt that both DPSG and Keurig viewed the cash represented by the Special Dividend as part of the consideration for the transaction. The DPSG proxy flatly states that “[t]he special cash dividend will not be paid to our stockholders unless the merger is completed.” The Proxy describes Keurig’s initial acquisition proposal in October 2017 as “an acquisition of DPSG in exchange for $66.00 per share in cash, payable in the form of a one-time [DPSG] dividend, and a 28% equity stake in the combined company to be retained by DPSG’s stockholders.” Additionally, DPSG’s financial advisors, in their fairness opinion on the transactions, considered the special cash dividend plus the retained DPSG Common to constitute the “DPSG consideration” received by DPSG’s stockholders.

B. Should a Distribution, Unaccompanied With an Exchange, Always Be Treated as a Dividend?

Recharacterizing corporate distributions for tax purposes is hardly a new phenomenon. As discussed below, there is strong precedent for treating certain exchanges as dividends, and certain dividends as exchanges, for tax purposes.

1. Recharacterizing an Exchange as a Dividend

The Code has several provisions that can recharacterize what is, in form, a stock sale as a dividend distribution for tax purposes. For example, when a corporation redeems its own stock, the sale will be recharacterized under section 302 as a distribution of property under section 301 by the redeeming corporation, unless the redemption satisfies one of three relevant tests showing that the redemption is not essentially equivalent to a dividend.

A redemption will be considered not essentially equivalent to a dividend under section 302 if, in general, the relevant shareholder suffers a meaningful reduction in its percentage stock ownership in the corporation as a result of the redemption and any related transactions. The complete termination of a stockholder’s interest in the corporation or a decline of more than 20% in a stockholder’s equity ownership (in each case subject to numerous additional conditions and qualifications and the application of section 318’s ownership attribution rules) provide safe harbors for sale or exchange treatment. In the context of a public shareholder in publicly-traded corporations, even a de minimis decline in that shareholder’s small percentage ownership is enough to qualify for sale or exchange treatment. In contrast, when such a shareholder’s ownership interest does not decline (or, in certain circumstances does not decline sufficiently) then the redemption will be considered dividend-equivalent.

Section 304 deals with a shareholder’s sale of stock in one corporation to another related corporation. This type of sale can be recharacterized as a dividend distribution to the selling shareholder in cases in which there is no meaningful decline in percentage ownership suffered by the selling shareholder (directly and indirectly) in the corporation whose stock was sold, after giving effect to the stock sale and any related transactions. Whether any decline in ownership is sufficiently meaningful to qualify for sale or exchange treatment is determined by reference to the dividend-equivalent tests outlined in section 302.

Section 356(a)(2) treats nonqualifying property (“boot”) received in a reorganization as gain from the sale or exchange of the surrendered property, except when the receipt of boot “has the effect of the distribution of a dividend.” Dividend-equivalence is once again tested by asking whether the shareholder in question suffered a meaningful decline in its percentage stock ownership of the resulting corporation, tested immediately after the reorganization and related transactions, determined by applying the section 302 tests for dividend-equivalence, adapted to reflect the two-corporation nature of the reorganization transaction.

Pursuant to the Supreme Court’s decision in Commissioner v. Clark, the specific test for the dividend-equivalence of boot in a reorganization assumes that the transferor shareholders receive acquiring corporation stock in lieu of the boot, and then have that hypothetical “extra” acquiring corporation stock redeemed for an amount equal to the boot actually received. If the transferor shareholder’s decline in its percentage stock ownership of the acquiring corporation resulting from this hypothetical redemption is sufficient under the tests developed in section 302, the boot is considered received in exchange for a portion of the shareholder’s acquired corporation stock; if the decline is not sufficient, the boot is considered to have the effect of a dividend and is taxed as such. For purposes of measuring E&P allocable to dividend-equivalent boot, it appears that only the transferor’s E&P is taken into account (although there is an argument that combined E&P should be used).

Finally, in Bazley v. Commissioner, the Supreme Court addressed whether the shareholders’ pro-rata exchange of par value stock in one corporation for an equal number of shares of no par value stock plus the corporation’s debenture was a “reorganization” or “recapitalization” of the corporation. Under then then-prevailing Code provisions, treatment as either a reorganization or recapitalization would have rendered receipt of the debentures tax free to the exchanging shareholders. In deciding that the transaction should be recharacterized as a (taxable) dividend of the debentures to the shareholders (rather than as consideration received for stock in the reorganization), the Supreme Court said:

In the case of a corporation which has undistributed earnings, the creation of new corporate obligations which are transferred to stockholders in relation to their former holdings, so as to produce, for all practical purposes, the same result as a distribution of cash earnings of equivalent value, cannot obtain tax immunity because cast in the form of a recapitalization-reorganization. . . . A “reorganization” which is merely a vehicle, however elaborate or elegant, for conveying earnings from accumulations to the stockholders is not a reorganization under § 112.

Even though the scope of Bazley is less well defined than the scope of the Code provisions discussed above, the precedents look through the “form” of a transaction “however elaborate or elegant” and apply federal tax law as if the transaction were “for all practical purposes” a different transaction, with a tax result more consistent with the underlying tax policies of the relevant statutes.

2. Recharacterizing a Dividend as an Exchange

When a corporation pays a dividend to its shareholders just before the corporation is acquired by another person, a series of precedents holds that the pre-sale dividend can be recharacterized as proceeds from the transfer of the target corporation’s stock to its new owner. Depending on the specific facts, the proceeds can be recharacterized as additional sales proceeds or boot in a reorganization.

The leading case in this area is Waterman Steamship Corporation v. Commissioner. There, a target corporation (Pan-Atlantic) was owned entirely by another corporation (Waterman). An unrelated corporation (McLean Securities) wished to acquire Pan-Atlantic for $3.5 million. If Pan-Atlantic were to sell its assets to McLean Securities and liquidate into Waterman under section 332, neither Pan-Atlantic nor Waterman would have recognized gain. Pan-Atlantic’s tax basis in its assets was at least $3.5 million, and Waterman’s $700,000 tax basis in its Pan-Atlantic stock would have been irrelevant under the nonrecognition provisions of section 332. Nevertheless, because of certain Interstate Commerce Commission (“ICC”) regulations, McLean Securities was advised by its attorneys to acquire Pan-Atlantic’s stock, and not its assets. Accordingly, the parties settled on the following plan, with all events occurring on January 21, 1955.

At noon on January 21, with ICC approval for the transaction having been obtained, Pan-Atlantic declared a $2.8 million dividend to persons who were stockholders of record (i.e., to Waterman) on January 20. While Pan-Atlantic had approximately $2.8 million of E&P—and based on the overall transaction had an equity value of $3.5 million—it had insufficient cash to fund the dividend, so the dividend took the form of a Pan-Atlantic promissory note payable one month after issuance. At 1:00 p.m., Waterman sold its Pan-Atlantic stock to McLean Securities for $700,180 pursuant to a contract of sale signed after the dividend was declared but before the sale of the Pan-Atlantic stock. At 1:30 p.m., McLean Securities and its principal shareholder collectively loaned Pan-Atlantic $2.8 million from funds they had borrowed from an unrelated party. Pan-Atlantic then immediately used the funds to pay off its $2.8 million dividend note.

By structuring the transaction as it did, under the then-prevailing Code and regulations, Waterman could receive the $2.8 million dividend both tax free and without reduction in the tax basis of its Pan-Atlantic stock, with the result that the entire $3.5 million was received tax free. In contrast, had Waterman simply sold the Pan-Atlantic stock to McLean Securities for $3.5 million, it would have recognized a $2.8 million gain. It is this latter view of the transaction that the Commissioner asserted in the litigation.

The Fifth Circuit held that:

The so-called dividend and sale were one transaction. The note was but one transitory step in a total, pre-arranged plan to sell the stock. We hold that in substance Pan-Atlantic neither declared nor paid a dividend to Waterman, but rather acted as a mere conduit for the payment of the purchase price to Waterman.

The court also noted, with apparent approval, the Commissioner’s argument that the funds needed to pay the dividend were supplied by the buyer of the stock.

Although Waterman Steamship arose in connection with the taxable sale of stock, its analysis has been expanded to apply to corporate distributions paid prior to, but in connection with, reorganization transactions. For example, in Revenue Ruling 75-360, based on the facts in McDonald v. Commissioner, the target corporation (E&M) entered into an agreement with an unrelated corporation (Borden) pursuant to which E&M would redeem preferred stock for $43,500 held by McDonald, its sole shareholder, with Borden then acquiring all of E&M’s common stock from McDonald for Borden stock. At Borden’s recommendation, E&M borrowed $43,500 on a short-term basis from a third-party trust company, redeemed the preferred stock, and a week later Borden acquired the E&M stock for Borden stock. Immediately thereafter, Borden contributed cash to E&M, which used the cash to repay the short-term loan. McDonald argued that the redemption should be considered a sale or exchange of the preferred stock (resulting in no gain to McDonald), with Borden’s acquisition of the E&M stock solely for Borden stock qualifying as a section 368(a)(1)(B) reorganization (“B reorganization”). The Service, on the other hand, apparently argued that the redemption and B reorganization were separate transactions. Viewed separately, the redemption was essentially equivalent to a dividend (McDonald owned all but a de minimis amount of E&M’s common stock after the redemption), and McDonald’s (full) basis in his preferred stock would be added to the tax basis in his Borden stock, instead of reducing gain recognized on the redemption of his E&M preferred stock.

Both the Tax Court’s decision and the Revenue Ruling are instructive. The Tax Court accepted the government’s stipulation that Borden’s acquisition of E&M common stock qualified as a B reorganization. But the Tax Court then integrated the acquisition with E&M’s redemption of McDonald’s preferred stock for purposes of testing the redemption for dividend-equivalence. As so integrated, McDonald’s equity ownership was reduced from 100% in E&M to a small minority interest in a large publicly-traded corporation. The section 302 tests for considering the redemption not essentially equivalent to a dividend were thus satisfied.

The Revenue Ruling states that the Service should not have stipulated that the transaction qualified as a B reorganization. Rather, it states that “since the acquisition [of E&M stock] was not solely for voting stock of the acquiring corporation but partly for cash, it is the position of the Service that the acquisition of stock of E&M did not constitute a reorganization.” The Service’s position thus assumes—similar to the court’s position in Waterman Steamship—that the cash McDonald received in redemption of his E&M preferred stock came from Borden, with E&M acting as a mere conduit, receiving cash from Borden and distributing it to McDonald for his preferred stock. So viewed, this cash consideration from Borden would have caused the transaction to fail as a B reorganization. This was the Service’s position even though the loan secured by E&M to redeem its preferred stock was neither supplied nor guaranteed by Borden.

The Service expanded this analysis in a Technical Advice Memorandum (“TAM”), discussing a transaction involving a dividend payment by a mutual insurance company (Target) to its policyholder owners. In that transaction, Target was to merge into another corporation (Taxpayer) in an A reorganization, immediately after Target paid an “extraordinary dividend” to its owners. The extraordinary dividend was intended to constitute a dividend to policyholders, deductible by Target under section 832(c)(11). Although the precise timing of the dividend declaration and payment mechanics are not entirely clear from the TAM, it is clear that the dividend was declared after the merger agreement was signed but before the merger date, payable to record holders of policies some number of days before the merger date, payable “on or about” the closing date, paid in connection with the merger, and actually distributed to policyholders via checks deposited in the mail within a short time after the merger. The dividend was contingent upon the fulfillment or waiver of all conditions precedent to Target’s merger with Taxpayer.

There is some dispute as to whether Target had sufficient capital for regulatory purposes to support payment of the special dividend. The TAM concludes that Target’s legal and financial ability to pay the extraordinary dividend is irrelevant, since Target did not in fact fund payment of the dividend:

Even if we accept Taxpayer’s assertion that Target had sufficient surplus to provide funding for the payment of the [extraordinary dividend] . . . we nonetheless conclude that having enough funds to pay a dividend is not dispositive of whether such payment is in fact a dividend. Rather, in this case the record shows that the [extraordinary dividend] . . . was a negotiated part of the consideration paid in the merger. . . . Target simply declared [the extraordinary dividend] . . . with Taxpayer furnishing the cash to make the purported . . . cash dividend. . . . The [extraordinary dividend] was declared as part of the merger, and it would not have been declared if all of the remaining steps in the transaction had not been lined up on the closing table and the merger did not in fact take place.

Based on this TAM and the other precedents discussed above, when a purported dividend is recharacterized as sale proceeds, the payment will give rise to additional gain (or reduced loss) to the receiving shareholders; if recharacterized as boot, then the boot will be tested for dividend-equivalency. If the receiving shareholders’ equity interest in the combined companies is reduced sufficiently, the boot will be considered gain from the sale or exchange of the shareholders’ stock. Interestingly, if the parties are forced by nontax considerations to have funds paid through a dividend distribution (rather than through a different, perhaps more direct form of payment), this may be relevant in determining whether the parties considered the payment a dividend or acquirer consideration, but is not dispositive as to the tax analysis outlined above.

3. Respecting Dividends and the Source-of-Funds Test

Just as some stock sales are not recharacterized as dividends, some dividends are not recharacterized as stock sales. This line of precedent respects pre-sale or pre-merger dividends as such for tax purposes. Revenue Ruling 69-443 nicely illustrates one aspect of these precedents. There, a corporation (Y) declared on November 1st its regular, annual year-end dividend to shareholders of record on November 1st, payable the following January 31st. On January 1st, X acquired Y’s stock solely in exchange for X voting stock. The Revenue Ruling held that the acquisition qualified as a B reorganization, despite the post-acquisition cash dividend distribution to Y’s (former) shareholders. This is because the dividend was paid by Y “out of its own funds” and was “not part of the consideration given by X in connection with the exchange of [Y stock for X] stock.”

Similarly, in Revenue Ruling 75-493, the Service found that a dividend of “unwanted cash” the day before an agreement to sell the stock of a corporation would be respected as such because the dividend did not reduce the price to be paid by the purchaser for the stock. Waterman Steamship was distinguished based on the fact that in Waterman Steamship “the funds to pay the dividend were actually furnished by the buyer.” And in Reitz v. Commissioner, the taxpayer was unsuccessful in his attempt to recharacterize a pre-gift dividend of cash and accounts receivable as sales proceeds, even though the dividend occurred a day before the stock (of a privately-owned hospital) was gifted to the local government and the dividend was referenced in the documentation. The Tax Court did not allow the taxpayer to question the form of the transaction as adopted by the taxpayer. Importantly, there was no evidence that the local authorities negotiated for a sale, participated in the dividend declaration, or would have paid anything to the taxpayer had the dividend not been declared.

Finally, in Litton Industries v. Commissioner, the Tax Court did not recharacterize a pre-sale dividend when the dividend (albeit in the form of a promissory note) was paid six months before the stock was sold. The record indicated that the stockholder did not announce its interest in selling the stock until two weeks following the dividend payment and pursued multiple alternative transactions after the announcement, including various IPO transactions. Based on the six-month time lag between dividend and sale and the lack of any settled understanding with a purchaser, the case was easily distinguished from Waterman Steamship. The corporation also had, in the Tax Court’s opinion, “sufficient revenue” to pay off the dividend note from sources other than a sale of the corporation’s stock, including a partial public offering or borrowing.

As outlined above, a number of factors distinguish cases in which dividends are recharacterized as sales proceeds from those cases in which dividends are respected as such. Nonetheless, the most important factor appears to be the supplier of the funds to make the distribution. When the distributing corporation supplies the cash to pay the distribution, the dividend will be respected as such. If the acquiring corporation, whether directly or indirectly, supplies the cash, the purported dividend will be recharacterized as sale proceeds or boot. In the latter situation, the distributing corporation is then considered to be a “mere conduit” for the transfer of cash from the acquiring corporation to the distributing corporation’s shareholders.

Implicit in this source-of-funds test is a question regarding the distributing corporation’s economic capacity to make the payment. If it is clear that the distributing corporation did not have the economic strength on a stand-alone basis to pay a dividend of the size actually paid, then the funds must have been supplied by the acquiring corporation—for if the distributing corporation could not fund the dividend, from where else could the funds have come? If the source-of-funds test recharacterized dividends as sales proceeds only in situations where, or to the extent, the distributing corporation lacked the financial wherewithal to make the distribution on a stand-alone basis, the analysis would rest on more substantive and less formalistic grounds. But the precedents have not applied the source-of-funds test so narrowly, instead applying a more expansive formulation to recharacterize dividends as sale proceeds where the source of funds can in fact be traced to the acquiring corporation, even where the distributing corporation could have—or even eventually did—prove to be the source of the payment.

4. Characterizing the Special Dividend

Were the precedents described above applied to the Special Dividend, there is little doubt that it would be treated not as a dividend distribution, but as proceeds from the redemption of DPSG stock. First, the Special Dividend was an integral part of the overall transaction, considered to be part of the consideration received by DPSG’s stockholders. Its payment was contingent upon consummation of the Merger and would not have been paid absent the Merger. Once the Merger occurred, the pre-Merger DPSG stockholders had a contractual right to the Special Dividend. The Special Dividend proceeds represented neither unwanted assets nor normal dividend practice.

Second, it is doubtful that DPSG had the economic muscle to pay the Special Dividend on a stand-alone basis. Prior to the Merger announcement, DPSG Common’s trading value was about $17.2 billion—a value below the $18.7 billion that DPSG would have been required to borrow to fund the Special Dividend. Generously using post-announcement trading values, the DPSG Common was worth about $21.6 billion, and DPSG had consolidated term debt of $4.1 billion (which was not refinanced in the Merger). Borrowing $18.7 billion for the Special Dividend in those circumstances would have raised DPSG’s total debt to $22.8 billion, and reduced its equity value to $2.9 billion, producing a relatively high debt-to-equity-ratio of about 7.9 to 1. Based on the above calculations, DPSG certainly could have borrowed some portion of the funds needed for the Special Dividend payment, but how much, and on what terms, is not at all clear.

Third, Keurig actually supplied all of the funds necessary to pay the Special Dividend. DPSG, in the words of Waterman Steamship, “acted as a mere conduit for the payment of the purchase price to” DPSG’s shareholders. The entire financing package was arranged, negotiated, and executed by Keurig. At least $9 billion of the Special Dividend was furnished by Keurig’s principal shareholder who (in connection with, and contingent upon, the closing of the transactions) contributed that amount as equity to Keurig, who then distributed the funds to DPSG the day of the Merger, and who then deposited the funds with Computerserve to help fund the Special Dividend. The remaining $9.7 billion came from borrowings arranged—and actually borrowed—by Keurig, with the borrowed funds delivered to DPSG as part of the Merger. It is hard to imagine a clearer example of a mere conduit than DPSG.

It is true that the $9.7 billion in loans became the obligation of the combined DPSG-Keurig group immediately after the Merger, so to some degree the Special Dividend could be considered to have been funded by DPSG. Under this more sympathetic analysis, based on the relative enterprise values of the two companies (without regard to the $9 billion equity infusion into Keurig which was immediately distributed), perhaps up to one-third of the Special Dividend could be considered to have been obtained or supported by DPSG’s pre-Merger business. But Waterman Steamship, and the precedents following it, hold that even when the distributing corporation could and does initially borrow the funds needed to pay the distribution, if the acquiring corporation actually funds or supports the borrowing, that fact alone is sufficient to consider the funds as having been supplied by the acquirer.

The case for treating the Special Dividend as something other than a dividend distribution by DPSG to its shareholders thus seems quite strong, but for one potentially important fact. In all the precedents discussed above, the shareholders who received the “dividend” also exchanged their shares in the distributing corporation for some other form of consideration. In Waterman Steamship, for example, post-dividend shares were sold for cash to the party supplying the funds for the first-step distribution. In the instant transaction the pre-Merger DPSG shareholders did take the necessary step of, in effect, voting in favor of a Merger that would severely dilute their equity interest in DPSG. But the shareholders neither exchanged any DPSG shares for cash or other property, nor had the terms of their DPSG stock altered as part of the transaction. Given this fact, what else could the Special Dividend be other than a dividend distribution?

The argument that the Special Dividend can be nothing other than a dividend for tax purposes may be enough under current law to carry the day. But it relies heavily—if not exclusively—on the form of the transaction. Certainly form plays a significant role in determining the tax consequences of various transactions under Subchapter C, sometimes rightfully so. But given the Code’s overall desire to distinguish dividend-equivalent transactions from nondividend-equivalent transactions, the consolidated return regulations’ desire to recast the identity of the acquiring and acquired corporations, and the case law’s desire to have tax consequences depend on the “substance” of the transaction, it would not be too difficult to extend the precedents discussed above to recharacterize payments like the Special Dividend as consideration received in exchange for DPSG stock.

A statement that a corporation’s pro-rata distribution of property to its shareholders can never be recharacterized as a sale or exchange is, of course, incorrect. Noncorporate shareholders receiving a pro-rata distribution in partial liquidation of the distributing corporation are treated, under section 304(b)(4), as having sold a portion of their shares, even when the shareholders surrender no shares in exchange for the distribution. The liquidating nature of the overall transaction overrides the formal, dividend-like nature of the distribution. Additionally, section 301(c) imposes sale or exchange treatment once the distributing corporation runs out of E&P and the shareholder runs out of tax basis in her shares. The shareholder has received cash in excess of both E&P and basis, which needs to be accounted for, but how? Sale or exchange treatment seems appropriate in such a case, as it reflects the general sense that the shareholder’s claim against corporate assets in excess of retained earnings has been converted into cash or other property.

5. Revisiting the Source-of-Funds Test

While the source-of-funds test generally guides the precedents distinguishing actual dividends from in-substance sales proceeds, the foundation for the test is not entirely satisfactory. If the distributing corporation has the stand-alone financial wherewithal to make the distribution, why should the tax consequences change if the acquiring corporation serves as the financier? Because money is fungible, so long as there are multiple sources for the money (i.e., the distributing corporation has the ability to make at least a portion of the payment), the particular source chosen to fund the payment should not be the focus of a substance-over-form test. A test based on tracing the source of funds to a particular source also quickly turns into a facts and circumstances test, in which a court must determine and weigh several factual elements that are either difficult to discern (e.g., a party’s ability to supply the funds, the parties’ intent to treat the dividend as “consideration” in the acquisition, the “business purpose” behind the transaction or the specific form of transaction adopted, whether the dividend represented “unwanted” assets, etc.) or subject to taxpayer manipulation (e.g., the choice of the specific source of funds used when multiple sources are available). If an explicit source-of-funds test were to be incorporated into the Code or regulations, the better approach would be to abandon any notion of tracing and instead consider the payment to have been funded proportionately between the corporations involved in the transaction based on relative available borrowing power.

A better approach would be to move away from the source-of-funds test and adopt a specific Code or regulatory provision that directly tests a corporate distribution like the Special Dividend for sale-equivalence, when the dividend is integrated with other transactional steps that affect the shareholders’ percentage ownership in the distributing corporation’s stock. The contours of such a provision are easily illustrated in the DPSG-Keurig transaction. Focusing on the historic DPSG shareholders, the overall effect of the transaction was to reduce their percentage ownership from 100% to 17%—a dramatic reduction in ownership. Had the transaction taken the form of a share redemption—even a “pro-rata” redemption of the shares held by historic DPSG shareholders—that redemption, once combined with the DPSG share issuances to Keurig’s shareholders, would not have been considered a dividend for tax purposes, but instead would have been treated as a redemption not essentially equivalent to a dividend under section 302.

Nor would a more specific sale-equivalence test be particularly difficult to fashion. In fact, a very similar test has already been developed. Under section 356, dividend-equivalence is tested by treating boot as if it were paid by the acquiring corporation to target shareholders in a hypothetical redemption of an amount of acquiring corporation stock that the target shareholders would have received absent payment of the boot. If a cash distribution, when combined with other transactional steps, did not result in a meaningful reduction in percentage ownership, then its payment would be respected as a dividend for tax purposes.

As applied to the DPSG-Keurig transaction, this approach would ask (1) what percentage of the post-Merger DPSG Common would the historic DPSG shareholders have owned absent the Special Dividend and (2) what was their actual percentage ownership in post-Merger DPSG, now taking into account the effects of the Special Dividend. The change in percentage ownership would then be tested against the section 302 tests for dividend-equivalence. Using certain reasonable assumptions regarding relative values, the historic DPSG shareholders would have owned approximately 52% of the combined company absent the Special Dividend, and their ownership interest would have then been reduced to 13% as a result of the Special Dividend—a reduction in percentage ownership clearly consistent under section 302 with treating the Special Dividend as having the effect of a sale or exchange of stock. Basis usage could be calculated using the percentage of stock considered redeemed in the transaction.

Creating a section 302-like provision applicable to sale-equivalent dividends would still require courts to determine if the dividend should be integrated with other transactional steps for determining the overall effect of the transaction, as occurs when section 302 proper, and other related Code provisions, are applied to a transaction. Such an approach would avoid many of the difficulties and artificiality of the source-of-funds test, and minimize consideration of the inherently factual issues discussed above that animate current precedents testing whether a dividend distribution should be recharacterized as either sales proceeds or boot.

Both the Code and taxpayers care considerably about whether a corporate distribution should be characterized as a dividend or as dividend-equivalent, not because dividend treatment is always the right or preferred answer, but because there are real economic differences between transactions that reflect the economics of a sale and transactions that reflect the economics of a dividend. As the DPSG-Keurig transaction shows, those differences can arise even when the taxpayers receiving the dividend do not dispose of their stock in a related transaction. The Code’s treatment of dividends should evolve so that payments delivered by way of a dividend can be recharacterized as sales proceeds, even if the shareholders receiving the dividend have their percentage interest reduced through dilution rather than through actual sale of their stock.

VII. Conclusion

The DPSG-Keurig transaction represents yet another battle in the enduring struggle between form and substance in United States tax law. The bare bones facts of the transaction—some would say the “substance” of the transaction—suggest that DPSG’s shareholders disposed of 87% of their DPSG stock in exchange for cash. The specific steps taken in the transaction—some would say the “form” of the transaction—suggest that DPSG’s shareholders retained all their DPSG Common and simply received a dividend from DPSG, whose E&P included the E&P of Keurig, the corporation that DPSG in form acquired. In this particular battle, the tax law should prefer the substance of the transaction over the specific form that was adopted.

Thanks to William Walsh for his research assistance. Thanks to Eric Sloan and David Weisbach for comments on an earlier version of this Article. All errors are mine. Facts relating to the transactions described herein are based on publicly-available information. The opinions expressed herein are solely those of the author and do not represent the opinion of any other person or organization. 

    Author