In an interesting recent paper, Yale professors Gary Gorton and Andrew Metrick compared the sub-prime meltdown to a conventional bank run, with the difference being that instead of banks, the shadow banks faced the wrath of investor retreat and capital flight on the back of margin calls and rising uncertainty. From this perspective, just as conventional banks hold and trade in fiat money, shadow banks issue and trade in privately created money (through repos, re-purchase agreements), whose value unlike that of fiat money is deeply vulnerable to market uncertainty. Therefore, it is all the more important that just as regular banks are regulated by strong controls to protect their investors, shadow banks need to be appropriately regulated to protect the buyers of its "money".
They attribute the emergence of the shadow banking system to a three-fold process that involved money-market mutual funds capturing retail deposits from traditional banks, securitization moving assets of traditional banks off their balance sheets, and repurchase agreements facilitating the use of securitized bonds in financial transactions as a form of money. This coupled with "evolution of the bankruptcy code that allows securitized bonds to be used as a form of privately created money in large financial transactions", triggered a boom in the securitization market.
As Gorton and Metrick argue, securitization appeared to create "information-insensitive debt" that could be freely exchanged by people who did not need to check into its quality. At its peak, the role of gatekeepers like the credit rating agencies took a backseat, as a widespread belief in the soundness of securitized instruments took hold. The challenge now is to effectively signal to investors that even senior tranches are not as safe as they seem and therefore investors need to be careful.
Underpinning the securitization market was the safe harbour rules that forbid the securities investors from having any right over the underlying assets (on which the securitization was done). Newly issued rules in the US seek to expand the scope of safe harbour and mandate much better disclosure of loan-level information, limits the number of tranches that can be created in a securitization, etc.
Gorton and Metrick propose the creation of a new category of financial institution, 'narrow funding banks' (NFBs), which would bring securitization under the regulatory umbrella and assure that all repos were backed by high-quality paper. Those narrow funding banks would be the only buyers of securitizations and other investors could buy notes issued by the new banks.
In other words, NFBs would become "the entities that transform asset-backed securities into government-overseen collateral", thereby seeking to ensure (through the more rigorous oversight) that repos are backed by higher quality collateral. They would intermediate as gatekeepers to bridge information asymmetry and signal the quality of assets, thereby mitigating the risks associated with investing in securities. About the underlying philosophy, they write,
"History has demonstrated two successful methods for the regulation of privately created money: strict guidelines on collateral (used to stabilize national bank notes in the 19th century), and government-guaranteed insurance (used to stabilize demand deposits in the 20th century). We propose the use of strict rules on collateral for both securitization and repo as the best approach for shadow banking, with compliance required in order to enjoy the safe-harbor from bankruptcy."
I am not sure whether the creation of exclusive NFBs to trade in securitized assets would alleviate the problems that bedevil the securitization market. The fundamental malaise was that the process of slicing and dicing reached a level of complexity that it became impossible to locate and price risk with any reasonable degree of approximation. Further, the gatekeepers who were supposed to bridge the information asymmetry between those peddling these products and their investors, the credit rating agencies, failed miserably to signal the quality of assets created with any reasonable level of accuracy. There is little in the proposals that would address these fundamental problems nor atleast disincentivize them.
Presumably, strict rules on collateral for both securitization and repurchase agreements, coupled with stringent monitoring of these new banks, would mitigate the risks associated with the conventional securitization market. But that does not lend much reassurance since, as aforementioned, most of these were ostensibly in place when the securitization market in sub-prime mortgages blew the bubble. And if "maintenance of this safe harbor is the incentive for agents to abide by the proposed rules", then the same (albeit less rigorous) should have been adequate to dissuade banks from peddling their alphabet soup of collateralized instruments and investors from faithfully gobbling them up.
Worryingly, there is every chance that these NFBs would emerge as the new set of "too big to fail" financial institutions, and much more dangerous since they would house only the opaque instruments. And in the final analysis, any regulatory arrangement that relies on more rigorous monitoring, without addressing the fundamental incentive mis-alignments that generate distortions and excesses, may not have much promise of success.