The impact of the Bankruptcy Abuse Prevention and Consumer Protection Act of 2005 repo ‘safe harbor’ provisions on investors

ABSTRACT

The Bankruptcy Abuse Prevention and Consumer Protection Act (BAPCPA) of 2005 significantly expanded the exemptions from the normal workings of the U.S. Bankruptcy Code. Using a large sample of U.S. banks, we study investors’ reaction to news about the promulgation of the BAPCPA repo ‘safe harbor’ provisions and the influence extending such exemptions to repos collateralized by riskier collateral had on equity market information asymmetry. We find a negative market reaction to news events about the promulgation of BAPCPA, which subsequent cross-sectional analysis suggests is at least partly driven by repo exposure. This finding suggests that investors perceived the increase in finance risk from the extension of the ‘safe harbor’ provisions as dominating the perceived gain from accessing cheaper finance. Further, we find that the promulgation of BAPCPA gave rise to increased information asymmetry for banks with repo exposure.

KEYWORDS:

• Repurchase agreements
• bankruptcy code
• safe harbor

JEL CLASSIFICATIONS:

• m41
• g21
• g32

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Justin Chircop http://orcid.org/0000-0002-5235-8077

Notes

1 As of August 2015, the following institutions were designated as primary dealers: Bank of Nova Scotia, New York Agency, BMO Capital Markets Corp., BNP Paribas Securities Corp., Barclays Capital Inc., Cantor Fitzgerald & Co., Citigroup Global Markets Inc., Credit Suisse Securities (USA) LLC, Daiwa Capital Markets America Inc., Deutsche Bank Securities Inc., Goldman, Sachs & Co., HSBC Securities (USA) Inc., Jefferies LLC, J.P. Morgan Securities LLC, Merrill Lynch, Pierce, Fenner & Smith Incorporated, Mizuho Securities USA Inc., Morgan Stanley & Co. LLC, Nomura Securities International, Inc., RBC Capital Markets, LLC, RBS Securities Inc., SG Americas Securities, LLC, TD Securities (USA) LLC and UBS Securities LLC.

2 The Bankruptcy Amendments and Federal Judgeship Act of 1984 extended ‘safe harbor’ provisions to repo agreements. This bill was signed into law on 10 July 1984.

3 Finance risk is sometimes referred to as roll-over risk. It is the risk that repo financing is not rolled over on maturity. Moreover, repos expose companies to counterparty risk. This risk refers to the possibility that the counterparty either fails to satisfy margin calls or fails to honour the agreement to repurchase the underlying assets.

4 We acknowledge that the ideal control group for our difference-in-differences analysis would be composed of banks to which all regulations except for the BAPCPA regulations relating to repo transactions apply. This control group does not exist since all US banks fall within the scope of the BAPCPA.

5 Sometimes provisions are inserted in the repurchase agreement which prohibits the sale and/or rehypothecation of the collateral to third parties.

6 SFAS 140, ‘Accounting for Transfers and Servicing of Financial Assets and Extinguishments of Liabilities’ was issued in September 2000. It was amended by SFAS 166 (FASB 2009) ‘Accounting for Transfers of Financial Assets an amendment of FASB Statement No. 140’ and more recently FASB issued Accounting Standard Updated (ASU) 2014–2011 Transfers and Servicing (Topic 860) (FASB 2014) ‘Repurchase-to Maturity Transactions, Repurchase Financings, and Disclosures’ which expand disclosures on repurchase agreements.

7 Bankruptcy Abuse Prevention and Consumer Protection Act of 2005, Pub. L. No. 109-8, §907, 119 Stat. 23, 171–172

8 Vide Table 1 for the major Congressional events in 2005 that led to the passage of the Bankruptcy Abuse and Consumer Protection Act (BAPCPA) of 2005.

9 The use of three-day event windows is well established in the literature. For example, Bushee and Goodman (2007); Krishnamurthy, Zhou, and Zhou (2006); Pinnuck (2005) and Ali et al. (2004) all use three-day time windows in their research design.

10 Another approach is to include net repurchase agreement (calculated as the difference between REPO and REV_REPO) instead of including REPO and REV_REPO separately in our model. While such an approach would allow us to measure the association between net repo exposure and market reaction, it would average out the magnitude and direction of the association between both repo measures and the market reaction.

11 Variable bhckb493 of the FRY-9C statements.

12 Variable bhck4340 of the FRY-9C statements.

13 Variable bhck7205 of the FRY-9C statements.

14 In untabulated results we obtain the same inferences when we run the analyses without winsorizing the data.

15 Further detail on this approach is provided in Sefcik and Thompson (1986).

16 Prior to data refinement that is based on the Sefcik and Thompson (1986) adjustment we have 459 unique bank holding companies. Thus, the use of the Sefcik and Thompson (1986) adjustment leads to the loss of 10 banks.

17 We recognize that due to limited data availability we are unable to control for the type of repo agreements the company engages into. As previously discussed it is only after ASU 2014–2011 became effective in 2015, that companies were required to disclose detailed information about the type of repo agreements they engage in.

18 Given that the choice of the beginning and end of sample period is ‘ad-hoc’ in robustness tests we test for the sensitivity of our results to this research design choice.

19 The economic significance is calculated as the coefficient on the interaction term (0.0019) divided by the mean SPR for the sample (0.00248).

20 This is a rough proxy of total holdings of mortgaged backed securities since this variable is only based on recognised mortgage backed securities. Banks might have other such securities held as collateral on repo and similar arrangements with counterparties and which are not recognised in the financial statements. Unless otherwise specified, these might be rehypothecated in other repo agreements. Information about such collateral is not publicly available.