ABI Commission to Study the Reform of Chapter 11
III. Background on the Commission and the Study Project 17
Virginia, two in New York, and one in Chicago. The Commission also held numerous subcommittee meetings in between each of these executive session retreats.
At each of the executive sessions, the Commission reviewed issues raised by witness testimony or examined in the research materials prepared by the Reporter and the advisory committees, including the advisory committee’s recommendations on the initial study topics posed by the Commission. Moreover, the reports of the international working group informed many of the Commission’s deliberations. From these discussions, the Commissioners worked to identify areas of potential reform that would, among other things, improve case efficiencies, enhance business rehabilitations and creditors’ recoveries, and resolve uncertainty or ambiguity in the current law.
During the three-year study process, and in connection with its deliberations, the Commission compiled and reviewed, among other materials, an extensive database of empirically based articles and working papers concerning different aspects of chapter 11 of the Bankruptcy Code. In reviewing all empirical data, including the data cited in this Report, the Commissioners were aware of the limitations that frequently impact chapter 11 data, including the following. First, endogeneity bias often is a problem in chapter 11 studies: (i) to the extent that omitted or unattainable information affects results (e.g., off-docket activity and negotiations; makeup of creditor body; talent or dynamics of management team; financial condition of debtor prior to and on petition date; impact of prepetition management decisions on company’s case; and, in some instances, economic or industry cycles), omitted variable bias occurs; and (ii) when the causal direction between two variables cannot be determined, simultaneity bias might occur (e.g., does the presence of private funds in a case make it more successful or do private funds invest in cases that are better positioned to be more successful).66 Second, selection bias can occur when subsets of available data sources are not randomly sampled or the pool from which the sample is drawn is not representative of the entire population (self-selection bias can also limit empirical survey studies). Third, coder bias and intercoder reliability can skew interpretation or results (e.g., if more than one coder is involved in the project, each may interpret the often subjective items on a chapter 11 docket in different ways, despite efforts to achieve an acceptable intercoder reliability rate). Fourth, data are limited and subjective: for example, it is difficult to define “success” in chapter 11; it is difficult to determine if a plan is a traditional stand-alone reorganization or a merger or a third party sale — they are all change of control events, and many datasets do not capture these nuances; and outside of public bondholders, it is difficult to determine recoveries in chapter 11 cases, particularly for smaller cases. Finally, because of the biases and limitations noted above, as well as others not discussed here, it might be difficult to establish strong claims of causality in empirical studies of chapter 11 cases. Nevertheless, the Commission reviewed empirical data from numerous sources and supporting a variety of different positions on the issues before it; it found all of the data informative, and it used the data in its overall consideration of all relevant factors.
The recommended principles set forth in this Report are the result of the Commission’s study and deliberative process. The Commissioners voted on each principle, and a principle was adopted as a Commission recommendation if it received support from two-thirds of the Commissioners voting, with 11 favorable votes being the minimum required for a principle being reported as a
66 See generally Michael R. Roberts & Toni M. Whited, Endogeneity in Empirical Corporate Finance, in Handbook of the Economics of Finance (2014) (discussing these issues with endogeniety, as well as measurement errors in that context).