Today I was merely in the audience (a question I asked aside) as 100 or so (!) NYU law students attended a panel on carried interests. Panel consisted of Vic Fleischer, Jon Talisman again for the defense, Cardozo law prof Mitch Engler, and economist Joel Slemrod, currently visiting at Columbia.
Vic gave the basic rundown of the issues. Talisman laid out his case a bit more fully this time than when I saw him at the panel in Washington a couple of weeks ago. Although he noted he was the only non-academic on the panel, it was actually classic first year law school type stuff, aka familiar legal reasoning by analogy. We all know A gets capital gain treatment, B is a little bit like A, C is not unlike B, D is not that far removed from C, and therefore they all should get capital gain treatment. Well done though not to me persuasive.
Talisman made a point in response to my question that I didn't feel I could answer there without unduly hogging the floor, what with other people waiting to ask questions. But it was the classic reasoning by analogy without (I would argue) adequate grounding. He noted that the proposed legislation gives ordinary income rather than capital gain treatment based on disproportion in the interests. E.g., I put in no cash but get 20% of the return as compensation for my services, and the proposed legislation makes this disproportion the ground for denying CG treatment.
Talisman gave the example: A and B both put cash in a partnership that develops shopping centers. Case 1, they participate equally, doing lots of work, and get a 50% return each, which unambiguously gets CG treatment under current law. Why should this change because A does a bit more work than B and thus gets 60-40. For that matter, why is A here different than if he did his own thing completely, blending a lot of labor income in developing the shopping centers with his own cash, and getting CG treatment for the whole thing. So what's the deal with disproportion being fatal to the CG result?
The answer relates to evidentiary problems in determining tax consequences. If, in the case of the solo developer of a shopping center, or the guy who spends lots of time on his stock trading and therefore gets an extra profit, we could impute the labor income, we probably should and would. But we can't - the evidence is assumed to be missing to do the imputed transaction here. Disproportion simply provides evidence that someone must be getting labor income, since why otherwise would they get a bigger share than is merited by the cash down alone. To say we shouldn't impute labor income when we have evidence of it, because we don't in various cases impute it due to the lack of clear evidence, would be rather silly. Why not then give me CG treatment on my labor income in teaching classes? It's merely a technicality that I didn't get to commingle it with some ordinary return on an asset.
Joel Slemrod made a nice analogy to "notches" in the rate structure, which would take too long to explain fully here, but the gist was that, when tax treatment is unavoidably discontinuous (i.e., one iota more CG-like, and the whole thing switches to getting CG rather than ordinary treatment), you want to find break points where people can't cluster just barely on the better side of the line. Having enough of your own money to invest versus needing other people's money is a convenient break point, in this sense, assuming one can police non-arm's length (or at least not generally available) and typically nonrecourse loans. So the analogy Talisman suggested fails here because it doesn't sufficiently suggest actual substitutability between structures.
Mitch Engler gave an analysis from his paper with Noel Cunningham, to the effect that the whole thing should be analyzed as an implicit loan. $10M fund, I as the general partner put in no cash but get a 20% profits interest, this is like making me an interest-free $2M loan. If the interest rate is 10%, the "real" transaction ostensibly had matching $200K payments of compensation from the LPs to me and an interest payment from me to them. Current year result: I have $200K net taxable income from the inclusion, due to rules limiting interest deductions.
Mitch (and Vic) called this the most accurate way to tax the deal, which I didn't necessarily see. One equally could see it as paying the GP $2M cash that he invests in the partnership - why think of this as "really" involving a loan of the value of the profits interest?
On alternative grounds, however, I saw this as an interesting solution. Say $2M is our best estimate of the value of what is given to the GP, because he has 20% of the profit interests in a $10M fund. (Admittedly, valuation may be more complicated. He may have to meet a hurdle rate, on the other hand suppose we expect an extraordinary return here, relative to the cash invested, due to the labor component.) Allowing the GP to defer the inclusion at a market interest rate, and making the LPs (who may be tax-exempt anyway) defer the $2M deduction at a market interest rate, is pretty much neutral compared to requiring current inclusion and deduction. So I am prepared to see the Engler-Cunningham solution as involving time value-neutral loans of tax liability between taxpayers and the government, even if imputing a loan between the parties does not especially resonate for me.