Nicholas L. Georgakopoulos, Bankruptcy Veil-piercing

• Publication year: 2011

BANKRUPTCY VEIL-PIERCING

Nicholas L. Georgakopoulos*

INTRODUCTION 472

1. THE ABUNDANCE OF CONTRACT PIERCING 473

2. ALTERNATIVE EXPLANATIONS OF THE ABUNDANCE OF CONTRACT PIERCING 475

   1. Do Product and Vicarious Liability Displace Tort Piercing? 475

   2. Does Subsidiary Insolvency Lead to a Multitude of Contract Breaches? 477

3. THE BANKRUPTCY PIERCING SAMPLE 478

   1. Sample Winnowing 478

   2. Fiduciary Perplexity 479

   3. Whose Veil? 480

   4. Non-consensual Void 481

   5. Bankruptcy Piercing Activity 481

4. MESSAGES FROM BANKRUPTCY 483

   1. Consensual Tort & Universal Piercing 483

   2. Piercing Non-Debtors’ Veils 484

CONCLUSION 487

APPENDIX 488

* Harold R. Woodard Professor of Law, Indiana University School of Law—Indianapolis. I wish to thank Daniel Cole, David Hoffman, Andy Klein, John Scheible, and participants in the 2010 meetings of the Midwestern Law and Economics Association, the Conference on Empirical Legal Studies, and ATINER for helpful comments. I urge you to send your comments to me at ngeorgak@iupui.edu.

INTRODUCTION

Bankruptcy veil piercing disputes have no true torts, and the primary target of piercing activity is not the debtor. The first finding contradicts the conventional wisdom that veil piercing primarily occurs in tort litigation. The second finding contradicts the notion that a failing debtor’s owners are the likeliest targets of piercing.

The scholarship focusing on the doctrine of corporate veil piercing is voluminous.1 Yet, the previous data largely does not accord with two presumptions of the legal academy—namely that piercing in tort should be virtually automatic (piercing would prevent limited liability from diluting the

incentives for care of tort law) and that contract disputes are highly unlikely to merit piercing (contracting parties should protect themselves against the risk of their counterparties’ insolvencies whereas piercing would erode their incentive to protect themselves). This Article reveals that bankruptcy piercing focuses exclusively on contract.

In a previous article, I have shown that piercing is eminently desirable in contract disputes as an improvement over the common law error doctrine.2 The previous article could have drawn inferential support from collective judicial intuition by merely pointing out that courts pierce in contract at a more than

trivial rate. However, the article’s cursory glance at the evidence suggested that courts and litigants demonstrate a bias in favor of piercing in contract disputes when compared to tort disputes. Traces of this bias also appear in other empirical scholarship about piercing. The research of Professor Robert Thompson shows an abundance of piercing in both contract and tort disputes,

and greater frequency of plaintiff success in contract piercing.3 Similarly,

Professors Hoffman and Boyd report that piercing occurs with greater probability in consensual relationships than non-consensual ones.4 A tort between strangers, of course, is the paradigmatic non-consensual source of litigation, but not the only one.

1. See Christina L. Boyd & David A. Hoffman, Disputing Limited Liability, 104 NW. U. L. REV. 853, 854

   n.10 (2010) (reporting 5,482 law review articles on the topic of veil piercing).

2. Nicholas L. Georgakopoulos, Contract-Centered Veil Piercing, 13 STAN. J.L. BUS. & FIN. 121, 130–31 (2007).

3. Robert B. Thompson, Piercing the Corporate Veil: An Empirical Study, 76 CORNELL L. REV. 1036,

   1058 (1991).

4. See Boyd & Hoffman, supra note 1, at 895 (finding that a plaintiff asserting a claim as a voluntary creditor is more likely to be successful on a veil piercing motion than in a case without a voluntary creditor).

   The Priest-Klein hypothesis5 about plaintiff success rates casts doubt on inferences from existing evidence because any differences in observed success rates may result from biased litigation and settlement incentives rather than any difference in underlying law and true likelihood of success.6 This Article

   empirically examines alternative explanations of the observed frequency of contract piercing.

   
   

   Part I updates the comparison of contract-to-tort ratios in veil piercing and beyond. Part II discusses alternative explanations for the surprising frequency of contract piercing, including the substitution of tort piercing with more lenient theories of liability. Part III explains the construction of the bankruptcy sample. Part IV discusses its observations. After the conclusion of the Article, the Appendix lists the bankruptcy opinions analyzed herein.

   
   

   1. THE ABUNDANCE OF CONTRACT PIERCING

      
      

      Piercing in contract is surprisingly frequent compared to piercing in tort. The comparison requires a baseline. The baseline is the universe of all contract and tort opinions. If the ratio of total contract opinions to total tort opinions is very different than the ratio of opinions about piercing in tort to those in contract, then either a different mechanism produces piercing disputes, or the litigants choose to pursue one more intensely when piercing.

      
      

      Indeed, the comparison shows a vast change from overall contract and tort to their mix in opinions about piercing. Tort becomes much less frequent among piercing opinions than it is overall. Whereas this could be consistent with the notion that tort plaintiffs do not pursue their piercing claims as intensely as contract plaintiffs—which would give inferential support to the idea that piercing in contract is easier—the ensuing analysis will show that the more likely answer is that tort and contract piercing disputes are generated differently than general disputes.

      
      

      The data for the comparison consist of the entirety of opinions concerning contract and tort from 1947 to 2010, inclusive. The source of the data is Westlaw. For consistency with prior research, contract and tort opinions were

      
      

5. George L. Priest & Benjamin Klein, The Selection of Disputes for Litigation, 13 J. LEGAL STUD. 1, 4–5 (1984) (hypothesizing that where the potential gains or losses from litigation are equal for both plaintiff and defendant, the rate of success for plaintiffs or appellants will tend toward 50% regardless of the substantive standard of law).

6. See, e.g., Steven Shavell, Any Frequency of Plaintiff Victory at Trial Is Possible, 25 J. LEGAL STUD.

   493 (1996).

   
   

   retrieved using Thompson’s search criteria.7 Thompson’s method also identifies opinions that mention piercing. This is a text search, pursuing either all forms of the verb “pierce” followed by “the corporate veil” or the phrase “disregard the corporate entity.”8

   
   

   The results of these searches are shown in Table 1, which also contains the result of the statistical test.9 The searches identified over 75,000 contract opinions, of which 693 mention piercing. The searches identified over 162,000 tort opinions, of which 394 mention piercing. The searches also identified opinions about both contract and tort by implication.10 These include over 5,000 opinions, of which 113 mention piercing.

   
   

   Notable is the fact that in the sample of all opinions, contract opinions occur less than half as frequently as tort opinions, whereas in the sample of opinions that mention piercing, contract opinions are almost twice as numerous as tort opinions. The statistical test confirms the difference with great statistical

   
   

7. See Thompson, supra note 3, at 1036 n.1. West identifies contract opinions with a single Key Number,

   1. Tort opinions are those that fall under four Key Numbers, 379, 272, 48av, and 48avi. The resulting searches need date range specifications narrower than 1947 to 2010 because they exceed the maximum number of opinions that Westlaw retrieves. All counts were made on January 25, 2011. As West corrects errors in opinions’ Key Numbers periodically, later searches will likely produce slightly different counts.

8. Accordingly, the search for contract opinions that mention piercing becomes “to(95) & (“pierc! the

   corporate veil” “disregard the corporate entity”) & date(aft 1946 & bef 2011).”

9. The statistical test is called the chi-test and is used to determine differences between data that fall into categories, like categories of opinions or outcomes of tosses of dice or coins (as opposed to data that consist of

   numerical values). The test requires that the comparison sample be aggregated and requires the calculation of the expected number of each category. In the case of the piercing sample, the test requires we calculate the fraction that piercing cases (in aggregate) are of all cases, and then apply that fraction (about .0049) to determine the expected number of contract piercing opinions, tort piercing opinions, and piercing opinions with both (about 372, 800, and 28, respectively). The chi-test (included in built-in spreadsheet software functions as CHITEST) compares those to the actual counts of the last column of Table 1 and computes the probability that the difference is due to chance rather than a different generative mechanism. See generally NICHOLAS L. GEORGAKOPOULOS, PRINCIPLES AND METHODS OF LAW AND ECONOMICS: BASIC TOOLS FOR NORMATIVE REASONING 294–97 (2005); DAVID FREEDMAN ET AL., STATISTICS 525–46 (3d ed. 1998).

10. The number of contract opinions includes some opinions on both contract and tort. Similarly, tort

    opinions include some on both. Summing those two and subtracting the number of opinions on either contract or tort gives the number of opinions that include both contract and tort. For example, if the contract opinions were five, three of which include tort, and tort opinions four, then the number of opinions on either contract or tort would be six. Summing the five contract opinions and the four tort opinions and subtracting the number of opinions about either gives three, the number of opinions about both (3=5+4-6).

    The count of contract opinions gives the sum of (i) only contract opinions and (ii) of opinions on both contract and tort: call that c + b. The count of torts gives (i) torts only and (ii) both, t + b. The count for either contract or tort gives c + t + b. Note that the sum of the counts for contracts and that for torts counts twice the opinions on both and can

    ...