The Journal of Financial Economic Policy published my paper called "Bank Capital Regulation, ABS-CDO Exposures and the Cost of Restoring Solvency" (
papers.ssrn.com/sol3/papers.…). It revises (and shortens) an earlier working paper called "Regulation, CDO Exposures, and Debt Guarantees through the Financial Crisis."
To understand the motivation, I've long been skeptical of "deregulation caused the crisis" claims. It's because people making this claim tend to overlook the lengthy regulatory notices that permitted the activities at the heart of the crisis in the first place. These regulatory notices reflect banker preferences to fund assets with debt rather than equity capital. The 1996 Market Risk Amendment (
govinfo.gov/content/pkg/FR-1…) gave large commercial bank holding companies (BHCs) incentives to hold ABS-CDO tranches in their trading book as it lowered required capital to almost zero. The 2001 Recourse Rule (
govinfo.gov/content/pkg/FR-2…) created incentives to hold the collateral for ABS-CDO deals by reducing capital requirements on the highest rated tranches in their banking book. In simpler economic terms, the Market Risk Amendment directly affected demand to hold ABS-CDO tranches. The Recourse Rule indirectly affected demand, through what Erel et al. (
papers.ssrn.com/sol3/papers.…) called the "Securitization Byproduct" effect, as ABS-CDO issuers had incentives to hold parts of their own and other banks' deals.
The market for ABS-CDOs emerged in 1998, after the finalized Market Risk Amendment but before the finalized Recourse Rule. In the paper, after constructing my own total global, US BHC and US investment bank ABS-CDO issuance series from Green Street's deal-level data, I find a structural break near the finalized Recourse Rule's public release date for US BHCs and globally, but not for US investment banks. That's not proof. But it's consistent with evidence from an earlier paper (
papers.ssrn.com/sol3/papers.…), in which I showed that BHCs with subsidiaries that commented on the 2001 Recourse Rule, on average increased holdings of the highest-rated tranches that could be used as collateral in ABS-CDO deals; those BHCs substituted away from holdings of lower-rated tranches. The control group on average kept holdings of the highest-rated tranches low and constant. In this paper, I also find that large BHCs with subsidiaries that commented on the Recourse Rule had spikes in estimated debt guarantees, used as a measure of the cost of restoring solvency, during the crisis, not before. The control group shows no changes in the cost of restoring solvency. Lastly, I find that the cost of restoring solvency has a large association with CDO tranche holdings; the associations with the original 2013 and revised 2019 measures of Volcker Rule trading asset measures and other control variables are relatively small.
I still conclude that the role of the ABS-CDO market in that crisis remains under-examined and under-appreciated in policy debates. Early on, popular writings such as Michael Osinski's 2009 New York Magazine piece (
nymag.com/news/business/5568…), the Afterword in Frank Partnoy's 2009 re-release of his book F.I.A.S.C.O. (
wwnorton.com/books/978039333…) and Michael Lewis's The End (
www-stat.wharton.upenn.edu/~…) clearly laid out why they mattered. Larry Cordell and coauthors (
papers.ssrn.com/sol3/papers.… and
papers.ssrn.com/sol3/papers.…) did fantastic technical work to reconstruct the deals to show why they lay at the heart of the crisis. While Cordell et al. touch on the regulatory angle, the link to regulation has been downplayed in policy debates.
After the crisis, the Financial Crisis Inquiry Commission Report (
govinfo.gov/content/pkg/GPO-…) devoted about a page each to the Market Risk Amendment and the Recourse Rule. Among many other things, Dodd-Frank called for changes to bank securitization activities, without acknowledging the regulatory changes that incentivized them in the first place - agencies could have just rolled back the parts of the Market Risk Amendment and Recourse Rule that lowered capital requirements. Rather than eliminating risk-weighting after the crisis, the Basel III regulatory framework became even more complex, as Jim Barth and I showed (
papers.ssrn.com/sol3/papers.…). We also showed (
papers.ssrn.com/sol3/papers.…) that a simpler regulatory capital framework has benefits that outweigh the costs. Risk-based capital regulation remains in place because the industry and rulemakers want it that way, but that comes with a high cost of doing business from all of the required compliance staff (e.g., accountants, lawyers and quants) and plenty of unintended consequences.