Bankruptcy Valuations: A Pair of Modest Proposals

I want to take up Michelle Harner's call for "innovation and new approaches to valuation". Valuation may well be the most important issue in bankruptcy, and it is also the issue that is least subject to meaningful judicial review. Imagine a Court of Appeals trying to parse through discounted cash flow models or what are proper comparables. The lack of meaningful appellate review makes it all the more important that we get valuation right. 

Our current valuation system is fundamentally adversarial.  Both sides put up evidence in the form of valuation compurgators experts, and the court then has to decide on a valuation, which need not be either of the experts'. The use of expert witnesses always has problems--the expert is employed by a litigant, which creates a concern about expert objectivity.  (I say this as someone who sometimes serves as an expert witness.) The normal solution in our litigation system is to have experts are paid for their time, not their opinions (which are their own) or the results in the case.

Unfortunately, and for reasons I do not understand, this civil procedure norm is not always the case with bankruptcy valuation experts. For example, in Chemtura, the Equity Committee's expert (UBS) received not ony a monthly fee, but a "transaction fee" that increased with higher returns to shareholders, which would, in turn, require higher valuations. UBS was, in effect, a party in interest in the bankruptcy because it's compensation turned on the valuation court the bankruptcy court accepted. Apparently such a compensation arrangement goes to credibility, not admissibility. This is not how things should work. 

What else could we do? One move would be some sort of market test.  Asset sales clearly get to a market test, although it's often a strange and distorted market depending on what bundle of assets are being sold, who can bid, etc. Even if one likes market tests in bankruptcy, however, they just aren't possible in many situations, such as for adequate protection valuations. So let me offer two possible alternatives.

First, why rely on experts retained by the parties?  Instead, have the court retain its own special master for valuation issues, paid for by the estate as an administrative expense. This would have two benefits. First, the cost would be for one expert, not two.  Second, the expert would not have the lurking conflicts of having been retained by one of the parties. It's not a perfect solution by any means, however, as there could still be methodological flaws in the valuation, but there wouldn't be the concern of self-serving valuations. Using a special master for valuation harkens back to the SEC advisory opinions on valuation in Chapter X of the Bankruptcy Act. No one thinks those worked very well (Atlas Pipeline is a posterchild case for this), but the problems were not so much about competence, as about how slow the SEC was. If the valuation were to be done under contract and deadline with a special master, the speed problem might well disappear.  

If you're not buying the special master approach, however, perhaps I can interest you in another modest proposal:  how about adopting the basic form of baseball salary arbitration to bankruptcy? (fwiw, there was a legislative proposal a few years ago to apply this model to interchange fee regulation.) 

My understanding of baseball salary arbitration is that both the player and the team make an offer. The arbitration is confined to pick between the two offers; the arbitrator is not allowed to split the baby and come up with its own valuation. The idea behind such a system (however well it works in practice) is to push the parties' valuations closer together because they are risk averse and hopefully encourage a settlement and thus avoid the need for a judicial valuation at all. Not sure how well this could be adapted to bankruptcy situations in which there could be multiple parties with different valuation positions. 

I'm curious to hear thoughts on these proposals and other ideas.  Comments are open.