Rationalizing the taxation of reorganizations and other corporate acquisitions.

The third, and perhaps most important consideration, is that it would be necessary to decide when a change in X's underlying assets during H's ownership of X's shares is sufficiently great that H should be deemed to have experienced a realization event. A 50% threshold has superficial appeal, but is there really so much difference between a shareholder whose indirectly owned assets have changed by 40% and one whose indirectly owned assets have changed by 50%, so that it makes sense to tax only the latter? I hardly think so. If it is not, then there are only two truly defensible positions: either H should be immediately (and therefore continuously) taxed on any change in X's assets, however small, or H should not be taxed on any change in X's assets, however large.

C. Extreme Options

As applied to corporate reorganizations and other acquisitions, these are the two extreme options: full recognition and no recognition. The first is essentially Brauner's proposal--repeal of the corporate reorganization rules--expanded for the sake of consistency so that shareholders of corporate parties to reorganizations (or other corporate acquisitions) are taxed not only if they experience a realization event, but even if they do not. The second is essentially the ALI's proposal: shareholders of corporate parties to reorganizations (or other corporate acquisitions) are not taxed whether or not they experience a realization event, unless they end up holding nonqualified consideration, such as cash.

Taking the full recognition proposal first, what are its strengths? Just one thing, I think: it might reduce the distortions that accompany income tax deferral, at least in the corporate context. Not only would human shareholders who previously relied on corporate reorganizations be forced to ante up tax on their unrealized gains, so too would human shareholders of the surviving corporations involved in such reorganizations.

However, I am skeptical that this benefit would be realized to any significant extent. Under a full recognition regime, I would expect to see a complete halt to explicit corporate acquisition activity. If in Acquisition #1 above, not only X's shareholder, H, but each and every one of GE's shareholders was forced to recognize all of their otherwise unrealized gain, GE would never acquire X. Rather, X and GE would seek transactions, short of a full-blown acquisition, that would give each party as much of what it wants as possible. Thus, X might sell its assets to GE (which might at least save GE's shareholders from taxation, depending on how the rules are written), X might lease a large fraction of its assets to GE, or X and GE might enter into some sort of partnership arrangement. Creative solutions would surely flourish.

If the benefits of full recognition are likely to be largely ephemeral, what about the detriments? These, I fear, would be all too real. First and foremost, the full recognition regime, expanded to be consistent, lacks an underlying theory. This is not, of course, Brauner's fault, but mine. Brauner has an underlying theory: he would tax all shareholder realization events. Period. The problem with this approach is not its lack of theoretical underpinning, but its lack of consistency. The form of any given corporate transaction would be massaged so that shareholders seeking to avoid recognition would simply hold their shares. Such shareholders would not experience a realization event, and so would continue to enjoy nonrecognition in spite of Brauner. It was to fix this consistency problem that I added deemed realization to Brauner: any shareholder of a corporation involved in a reorganization-like transaction is deemed to experience a realization event, whether or not he continues to own his original shares. No nonrecognition provision is available in the event of either an actual or a deemed realization.

Unfortunately, once deemed realization is added to the mix, all of the theoretical problems discussed in the context of the intermediate taxation options emerge. In particular, there would be no theoretical reason to single out stock-based corporate acquisitions: realization would be equally appropriate for shareholders of corporations that acquired other corporations for cash, acquired such corporations' assets for cash, or acquired noncorporate assets for cash. There is no defensible stopping point on this slippery slope, short of taxing shareholders each and every time that the corporation exchanged any asset at all. Such taxation is continuous taxation, or mark-to-market taxation. While it is quite defensible in theory, it is well beyond the extreme option envisioned by Brauner.

In addition to the lack of a theoretical underpinning, or even in the presence of such theoretical underpinning (as would be the case if full recognition were deemed to be synonymous with mark-to-market), full recognition taxation carries with it a host of problems. I count at least five, all of which have been elaborated to a greater or lesser degree above: (1) discrimination against corporate transactions in favor of noncorporate transactions, assuming that nonrecognition rules in noncorporate contexts remain unchanged, (2) probable discrimination against public company corporate transactions in favor of private company corporate transactions, assuming that full recognition is either inapplicable or differently applicable in private company contexts due to the absence of readily available valuation data, (3) taxation of shareholders in the absence of cash receipts and perhaps even in the absence of liquidity, (4) possible taxation of shareholders in the absence of objective valuation data, and (5) taxation of shareholders in the absence of any voluntary transaction and in certain circumstances in the absence of any transaction at all, voluntary or otherwise (as would be the case if recognition applied to shareholders of both the target and the acquiring corporation, as would be demanded by consistency).

Turning at last to the only proposal still standing, the no recognition proposal, what can be said in its defense? Actually quite a lot: in particular, it does not suffer from any of the infirmities already elaborated for either an intermediate recognition regime or a full recognition regime. Moreover, it is breathtakingly simple. Finally, it is unlikely to meaningfully increase the tax-regime-based inefficiency of shareholder behavior, also known as the "lock-in" effect. The lion's share of such inefficiency is a direct result of the realization principle in its naked glory. Furthermore, neither incrementally reducing the effect of such principle by making nonrecognition randomly (from a shareholder's perspective) inapplicable to certain corporate mergers and acquisitions nor incrementally increasing its effects by making nonrecognition applicable to a broader array of corporate mergers and acquisitions is likely to have any meaningful impact on the underlying problem. (78)

Alas, once one concludes that expanded availability of nonrecognition treatment is the only defensible change to the taxation of corporate mergers and acquisitions, it is necessary to decide where to draw the line on nonrecognition. Should one, as the ALI proposed, draw the line at the receipt of consideration other than stock, making any such receipt taxable? I think not. I think that inherent in the realization principle is the notion that it should be the taxpayer's own choice that determines when he will be taxed on his gain with respect to any single investment. Thus, if a conceded realization event occurs that is not of the shareholder's choosing and, moreover, is against his wishes, such realization event should not result in the immediate payment of tax. Congress has recognized this principle in noncorporate contexts: section 1033 applies to "involuntary conversions" of property, but that provision does not include within the definition of "involuntary conversion" the "forced sale or exchange" of corporate stock pursuant to a merger or acquisition. (79) I think that the involuntary conversion rules, or something very similar, but with a considerably shorter reinvestment window to reflect the fact that finding replacement property in the corporate equity context is generally easier than finding replacement property in the context of other types of property, should be made applicable to corporate mergers and acquisitions.

This suggestion has been made once before, albeit as a throw-away remark. (80) That remark has been criticized by Brauner, who correctly notes that the possibility of a merger or acquisition is inherent in any shareholder's stock, and that there is accordingly no unfairness in taxing such shareholder as and when a merger or acquisition arises, regardless of whether the shareholder has consented. (81) But Brauner's argument proves too much. The possibility of destruction by fire or confiscation in an eminent domain proceeding is inherent in many types of property, but that does not make such property ineligible for nonrecognition relief under section 1033. Thus, the premise behind the current involuntary conversion provision can be none other than that it is unfair to remove from the taxpayer the unilateral right to determine when he will recognize his gain. Period. This argument has as much force when the property is common stock as when it is real property or livestock. (82)