The Treasury plans to invest up to $250 billion to recapitalize the U.S. banking system by buying non-voting preferred stock and warrants to purchase common stock from a large number of banks, thrifts, bank holding companies and savings and loan holding companies. The Treasury has caused guidance (Notice 2008-100) to be issued confirming that Treasury’s preferred stock and warrant investments are to be disregarded for certain purposes of section 382, which controls net operating losses.
As useful as that guidance may be, a question has arisen: What are the implications of Treasury’s non-voting preferred stock for future tax-free acquisitions of those banks, thrifts and holding companies, while the non-voting preferred stock is outstanding (whether in the hands of the Treasury or others)?
Corporation T has two classes of stock outstanding: voting common, worth 90 percent of its total equity, and non-voting preferred issued initially to Treasury, worth 10 percent of its equity. Corporation X wants to acquire all of the stock of T for X stock. T wants its common shareholders to receive common stock of X and wants its preferred shareholders to receive preferred stock of X with the same characteristics—i.e., non-voting. Both parties want T to remain in existence. It is not feasible for X to acquire all of the T stock through a public tender offer; therefore, T and X agree that X will cause a transitory subsidiary that it will create for this purpose, Mergerco, to merge into T and the T shareholders will exchange their T shares for the described two classes of X stock. Thus X will acquire all of the T voting stock for X voting stock and will acquire all of the T non-voting stock for X non-voting stock. It is not certain that this transaction can qualify as a tax-free reorganization.
The “reverse subsidiary (or triangular) merger,” as it is called, can qualify as a tax-free (actually tax-deferred) reorganization under the IRC, either under section 368(a)(2)(E) or section 368(a)(1)(B). It normally would be thought to qualify under the former section, which was added to the IRC in 1971 specifically to allow the reverse subsidiary merger to qualify as a reorganization in a way other than a section 368(a)(1)(B) acquisition of stock “solely for voting stock.”
However, the relaxation of the prohibition against “boot in a B” only went so far. Specifically, section 368(a)(2)(E) requires that “…in the transaction, former shareholders of the surviving corporation [here T] exchanged, for an amount of voting stock of the controlling corporation, an amount of stock in the surviving corporation which constitutes control of such corporation.” The problem arises from the way the IRC defines “control” for this purpose. Section 368(c) has long been interpreted to define control not as 80 percent of aggregate vote and value, but as 80 percent of the vote and 80 percent of each other non-voting class (without aggregating the non-voting classes).
This appears to mean that X would have to acquire at least 80 percent of T’s non-voting preferred stock for X voting stock, be it voting common or voting preferred. See Bittker & Eustice at fn. 287 ¶12.25[a], indicating such a reading may be appropriate.
There are other supporting indications. The legislative history of the 1971 adoption of section 368(a)(2)(E) states
SECOND, IN THE TRANSACTION, FORMER SHAREHOLDERS OF THE SURVIVING CORPORATION MUST RECEIVE VOTING STOCK OF THE CONTROLLING CORPORATION IN EXCHANGE FOR AN AMOUNT OF STOCK REPRESENTING CONTROL IN THE SURVIVING CORPORATION. CONTROL FOR THIS PURPOSE (DEFINED IN SEC. 368(C)) MEANS THAT THE AMOUNT OF STOCK IN THE SURVIVING CORPORATION SURRENDERED FOR VOTING STOCK OF THE CONTROLLING CORPORATION MUST REPRESENT STOCK POSSESSING AT LEAST 80 PERCENT OF THE TOTAL COMBINED VOTING POWER (IN THE SURVIVING CORPORATION), AND ALSO STOCK AMOUNTING TO AT LEAST 80 PERCENT OF THE TOTAL NUMBER OF SHARES OF ALL OTHER CLASSES OF STOCK (IN THE SURVIVING CORPORATION). IF VOTING STOCK OF THE CONTROLLING CORPORATION IS USED IN THE EXCHANGE TO THE EXTENT DESCRIBED, ADDITIONAL STOCK IN THE SURVIVING CORPORATION MAY BE ACQUIRED FOR CASH OR OTHER PROPERTY (WHETHER OR NOT FROM THE SHAREHOLDERS WHO RECEIVED VOTING STOCK). (emphasis added) S. Rep. 91-1533 to P.L. 91-693.
Also, Reg. section 1.368-2(j)(6) Ex. 3 deals with T having voting and non-voting stock. The example does not state the value of the two classes, but only that there are 1000 common shares and 100 non-voting preferred shares; before the merger, T redeems the preferred shares for its cash. The example states that such shares will not be considered to be outstanding immediately before the merger, and therefore X will acquire control of T entirely for X stock through a reverse subsidiary merger in which the sole consideration is X common stock.
This example implies that if the preliminary redemption had not occurred, and if the preferred stock had been cashed out in the merger, then the merger would not have qualified under section 368(a)(2)(E), because part of the acquisition of control, the acquisition of at least 80 percent of the non-voting class, would have occurred in exchange for boot and not X voting stock. Indeed, the example probably was intended to show a way out of that problem, and to provide a safe harbor for such a technique, which otherwise might be problematic.
Of course arguments can be made against the interpretation suggested above. An example involves section 368(a)(1)(B), under which a reverse subsidiary merger might also qualify as a tax-free reorganization, if “solely for voting stock.”
In the 1970s, the question arose as to whether that requirement was satisfied if 80 percent of T’s single class of stock was acquired for voting stock and some boot was paid for additional shares. The tax court ruled that voting stock had to be paid only for the 80 percent in Reeves, 71 TC 727, but reversed in Heverly, 621 F. 2d 1227 (3d Cir. 1980). The Tax Court approach, however, was followed in Pierson v. U.S., 472 F. Supp. 957 (DC Del. 1979).
The positive analogy that can be drawn from the appellate decision in Heverly is that, for purposes of the B reorganization, the court looked at the totality of the acquisition of the T stock, and not just at the acquisition of the control part. That would be helpful in the reverse subsidiary merger case, if one could devise a method of applying the statutory requirement to the aggregate acquisition of T stock.
However, that brings one back to the basic problem: There seems to be no way to test the control acquisition other than as a two-part acquisition: acquisition of the controlling 80 percent piece of the vote and the controlling 80 percent piece of non-voting classes. There appears to be no way to shift, for example, to a requirement that 80 percent of the T vote and value be acquired for voting stock, because that just is not the way the statute defines control: it has no value element.
There are many ways out of the problem described above. The acquirer can give voting stock only. T can recap the non-voting preferred stock to voting preferred stock before the merger and it can be exchanged for X voting preferred stock. As the regulation example illustrates, T can redeem the preferred stock before the merger. Furthermore, in light of the leniency currently being exercised by the Treasury in other areas of the corporate tax laws applicable to distressed banks, it is not beyond imagination that the Treasury would simply issue guidance to the effect that the Treasury stock would be ignored for purposes of qualifying as a reorganization. See Notice 2008-83 providing favorable guidance concerning banks’ built in losses under section 382.
TARP Capital Purchase Program Term Sheet. Some of the planning will be affected by the proposed terms of the senior preferred to be bought by Treasury. The preferred cannot be redeemed for three years, except with the proceeds of an offering of replacement Tier 1 capital. The dividend rate will rise from five percent to nine percent after five years, which would suggest that the Treasury expects to be divested by that time, or to encourage the banks to facilitate such divestment by that time. The Treasury can transfer the stock, which means that the Treasury as a cooperative force in a reverse subsidiary merger could be out of the picture after a period of time.
Assuming that considerable bank consolidation will take place within the three-year period (and probably a lot quicker than that), it is likely that Treasury will hold the preferred at the time of some proposed acquisitions by reverse subsidiary merger, and that the banks may be unable to raise sufficient replacement Tier 1 capital in order to redeem the preferred. This means that Treasury either will have to agree to take voting preferred of an acquirer or to waive the limitation on redemption, or to provide other relief administratively.
If the preferred stock needs to be converted to voting preferred stock or acquired for similar voting preferred stock, a change may be needed in Notice 2008-100, III-c, which provides that, “for all federal income tax purposes,” the preferred stock will be treated as section 1504(a)(4) stock. That section requires non-voting stock, which the Treasury’s stock will be initially. A later addition of voting rights, however, would remove the stock from section 1504(a)(4) as normally applied, and might be inconsistent with the premise under which the Notice treats the stock as conforming to that section.
Further consolidation of the banking industry will occur, and sooner rather than later. If the recipients of Treasury capital injections are acquirers, the problem above will not arise (although other issues, such as the protection of the seniority of the Treasury preferred, will have to be addressed). If the recipients of Treasury’s capital are the targets, however, then the problem is likely to occur in some cases.