Bankruptcy contracting reviewed.

• Author: Schwartz, Alan
• Position: Response to article by Lynn M. LoPucki in this issue, p. 317

Western bankruptcy systems have two relevant features: (a) The systems are mandatory, that is, parties are required to use the state supplied system; and (b) each system has a number of mandatory rules. Until recently, reformers took these features as givens. The reformers' goal has been to improve the mandatory bankruptcy system and the mandatory rules within it.(1) Some scholars now contest the first relevant feature, arguing that requiring parties always to use one bankruptcy system is inefficient. In a recent essay,(2) challenged both of these features.

I made these claims:

(A) The only goal of a business bankruptcy law should be to reduce the cost of debt capital, which the law best does by maximizing the debt investors' insolvency state payoff.

(B) Regarding mandatory bankruptcy systems:

(i) Requiring parties always to use the mandatory state system increases a borrowing firm's cost of capital over the cost that would obtain in a world in which the firm and its creditors could contract for an alternative bankruptcy system.

(ii) If the rule against contracting for a preferred bankruptcy system were relaxed, parties would write "bankruptcy contracts" that would induce a borrowing firm to choose the system that would be optimal for it and its creditors were it to become insolvent.

(C) There should be few mandatory rules within bankruptcy systems. More precisely, [sections] 365(e) of the Bankruptcy Code, which prohibits the use of ipso facto clauses,(3) should be repealed. An analysis of this section also suggests that the wisdom of other mandatory rules should be rethought.

Professor LoPucki's reply(4) to my essay does not challenge claims (A) and (C) and explicitly accepts claim (B)(i). Professor LoPucki attempts to refute claim B(ii), and he makes a new, related claim. Regarding his refutation, I showed formally that parties would write bankruptcy contracts when all creditors (a) lend to the firm at the same time and (b) have the same preferences regarding bankruptcy systems. I then extended this result to argue that (a) parties would write these contracts when creditors lend at different times; and (b) creditor conflict regarding bankruptcy contracts would be rare if courts followed the absolute priority rule, and the few inefficient refusals to sign these contracts would be eliminated were the law to permit the preferences of a majority (in amount) of creditor claims to determine whether a bankruptcy contract became effective. My essay thus predicted that parties would write bankruptcy contracts if they were free to do so and if bankruptcy law were appropriately modified. This prediction is not testable today because bankruptcy contracts are unenforceable.(5) Hence, my essay attempted to motivate reform. If there is a possibility that free contracting over bankruptcy systems would increase welfare, and if there otherwise is nothing wrong with free contracting, then free contracting should be permitted.

Professor LoPucki argues that, even if one accepts the assumptions of my model, strategic behavior by borrowing firms would prevent the writing of bankruptcy contracts when the firm's creditors have the same preferences regarding bankruptcy procedures but lend at different times. He goes on to argue that actual creditors' preferences regarding bankruptcy systems almost always differ, so that creditors could not agree on a bankruptcy contract even if they lent at the same time. Taken together, these arguments lead Professor LoPucki to conclude that bankruptcy contracts seldom would arise if the prohibition on writing them were reversed.

This conclusion, if true, would be helpfully clarifying but would not count strongly against an argument for greater freedom of contract in the bankruptcy field. If free contracting over bankruptcy systems would be efficient in ideal conditions, then the fact that scholars today cannot plausibly show how real parties would write real contracts is not a serious refutation. The real parties should be given a chance to see how much of the scholar's heaven they can actually enter. Professor LoPucki's new claim responds to this point. In his view, sophisticated parties today exploit current law to write contracts regarding bankruptcy that redistribute wealth to themselves from less sophisticated parties and involuntary creditors such as tort victims. Parties would, he claims, use more freedom to contract to engage in greater redistributional efforts. The relation between Professor LoPucki's new claim and his challenge to my argument is this: If current contracting practices support an inference that parties would abuse a greater right to contract, giving them this right is unwise. A showing that my model does not work thus both serves to defeat my particular claim about bankruptcy contracting and helps to impeach the more general claim that bans on free contracting should be repealed.

I am grateful to Professor LoPucki for giving me the opportunity to clarify certain steps in my argument, but his reply itself is unsuccessful. A disagreement between scholars can take three forms: First, there can be disagreement about the state of the world. In law and economics, this disagreement commonly takes the form of contesting the assumptions that underlie a model. Second, there can be disagreement about the norms that do or should apply to the case under study. Third, there can be disagreement about the analysis. In this third category, the issue is whether a scholar's conclusions follow from his premises. Professor LoPucki's critique takes the third analytic form. He argues that my premises entail less favorable conclusions for bankruptcy contracting than I claim. This critique fails, however, because Professor LoPucki's arguments are either ruled out by my model's plausible assumptions or reflect a misunderstanding of the model. It does not follow that my policy prescriptions must be accepted, because there still can be disagreement over facts or norms. But since Professor LoPucki does not argue about facts or norms, and since my model withstands his analytical critique, we are left where my essay ended. In the absence of a plausible disagreement with my analysis that takes the first or second form, the state should give free contracting a chance.

Part I below refutes Professor LoPucki's criticisms of my model. Part II shows that current contracting practices are irrelevant to the question whether ex ante bankruptcy contracts should be lawful, and thus that these contracts cannot support an inference that greater freedom to contract about bankruptcy would be harmful. Part III is a conclusion that briefly discusses what may be the basic source of the disagreement between Professor LoPucki and me: that bankruptcy in my view is, and apparently in his view is not, a part of the law of business transactions generally.

1. BANKRUPTCY CONTRACTING

   1. When Creditors Lend at Different Times

      1. A Summary of the Model

         My model assumed that two bankruptcy systems existed. One system, denoted L, resembled the current Chapter 7, and the other, denoted R, resembled the current Chapter 11. For the purposes of this Response, the parties in my model could write two types of bankruptcy contracts. The first would not deal explicitly with bankruptcy at all, leaving the insolvent firm free to choose the bankruptcy system it preferred ex post. Since creditors are legally entitled to the full monetary return from a bankruptcy procedure, the firm would not consider this return in making its choice. Rather, the firm's owners/managers(6) would choose the bankruptcy system that maximized their private benefits. The parties, however, could renegotiate after insolvency to induce the firm to choose the bankruptcy system that generated the highest monetary return when that system did not also maximize the firm's private benefits. The model assumed that a firm always gets greater private benefits in the reorganization system R because the managers get to run the firm for a longer time in that system and also have some chance of saving the business. Hence, the parties would renegotiate only when the liquidation system L turned out to generate a higher monetary return than the reorganization system R. The creditors then would pay the firm a sum to forgo the R system's greater private benefits and instead enter liquidation.

         The second contract I discussed--and the one Professor LoPucki considers--authorized the firm to keep a portion of the monetary return that would be generated by whatever bankruptcy system it chose. If this portion--the "bribe"--were set appropriately, the sum of the private benefits and cash payments the firm would get if it chose the optimal system always would exceed the firm's total payoff from choosing suboptimally. This contract was called renegotiation-proof because the parties would have no need to renegotiate later: The firm would choose the efficient bankruptcy system if the contract bribe was correctly specified.

         One or the other of these contracts would maximize the creditors' insolvency-state payoff and thereby minimize the firm's cost of capital, depending on the relevant economic parameters. For example, the renegotiation contract would often be best when the reorganization system had a higher expected return than the liquidation system. The creditors do not pay a bribe under this contract and thus could keep the entire high return that system R would generate; the firm would choose R without a bribe because it would get greater private benefits in that system. If the liquidation system generated the higher expected return, on the other hand, the renegotiation-proof contract often would be best. The firm then would have to be bribed to choose liquidation. The ex post bribe would exceed the contractual bribe because ex post it is known with certainty that the firm will choose the inefficient system; hence, the firm could exploit its bargaining power to capture most of the...