Recent regulatory proposals advance many changes to the structure of the financial services industry. The chief economic problem with those proposals, however, is that many of them go against the grain of decades of economic research on financial system design and incentive conflicts.
Voluminous academic literature on bankruptcy costs, deposit insurance, banking and regulatory policy provides templates for reform. Virtually none of the lessons from that literature are acknowledged in the proposals.
Take, for instance, bankruptcy costs. While it is fine to advocate a resolution regime for nonbanks, the fact is that deeply insolvent firms are costlier and more difficult to resolve (see, for instance, Weiss 1990, Journal of Financial Economics). We saw this most recently in the thrift crisis, where regulatory forbearance left managers to "bet the bank," leaving entanglements that were costly to sort out after managers had carted off most of the cash.
After the crisis, the FDICIA's prompt corrective action provisions were put in place to prevent such forbearance and the recurrence of unnecessary losses. The idea was to shut banks down when they had remaining tangible capital so that the excess value could cover resolution costs. The positive capital would also make resolution easier when going-concern value was preserved. Regardless of how well Tier 1 financial holding companies plan their own funerals, the proposed policies do nothing to alleviate Tier 1 FHCs' systemic (and political) importance that will prevent them from being closed well in advance of insolvency, in contrast to the proper economic intent of PCA.
Such influences further undermine the rules-based approach of FDICIA's PCA provisions, which were obviated again in the present crisis when regulators refused to downgrade bank Camels ratings to levels at which FDIC resolution authority takes hold. Despite a substantial body of economic literature on the inefficiency of discretionary regulatory principles relative to well-designed and well-articulated rules (see, for instance, Kydland and Prescott, 1977, Journal of Political Economy), the Obama administration seeks not to enforce and build upon existing rules but to ignore them altogether.
That's OK, says the administration. We will assess the Tier 1 FHCs to cover the additional costs. The problem with the approach, however, is that only a subset of institutions is assessed and members of that subset influence their inclusion in the assessed class — whether by virtue of size, "systemic importance," or other factors. Well-managed and solvent institutions will be able to use their influence or control to stay outside the penalized subset of institutions. Hence, the only institutions assessed at the higher rate will be those that lack the wherewithal to wield such influence or control, and are therefore doomed to fail.
Again, this is nothing new. Deposit insurance experiments in the U.S. in the 1800s and early 1900s also allowed voluntary participation. Time and again, those deposit insurance schemes failed when the only banks that chose membership were those about to fail (see, for instance, Calomiris 1989, Federal Reserve Bank of Chicago "Economic Perspectives"). Without assessments on solvent institutions, those deposit insurance schemes soon went bankrupt. We can reinvent the wheel here, but the FDIC is already assessing all institutions on the basis of assets rather than deposits, meeting the policy objective of charging Tier 1 FHCs for their systemic importance.
At the end of the day, the administration remains wedded to supervising directly and providing a safety net for all financial institutions, not just depository institutions. But a vast academic literature is built upon the basis that nondepository institutions are not "special" enough for such subsidies, because they do not provide significant value to society or the economy.
The standard view (see, for instance, Bernanke and Gertler, "Banking and Macroeconomic Equilibrium" in "New Approaches to Monetary Economics," 1987) is that depository institutions serve small unsophisticated investors (depositors), helping to use their funds for investment purposes — typically small business and consumer loans — and in turn stimulating economic growth. In doing so, depository institutions bridge substantial information gaps in ways that cannot be replicated easily. Brokers, investment banks and other nondepository institutions stand between sophisticated investors and well-known seasoned firms, intermediation functions that can be replicated relatively easily. Hence, depository institutions are uniquely protected by safety net subsidies to preserve their valuable economic and social functions.
Today's problem is the "interconnection" between the depository and nondepository institutions. But while those connections are still supposed to be limited under Gramm-Leach-Bliley, they have been eroded recently by waiving Federal Reserve Regulation 23a in the crisis and they will be eroded further still by the recently advanced regulatory proposals.
The proposal of a single financial regulator seems suggest the devolution to a single type of financial institution. If that is the policy direction that is desired, let's say so. Then it becomes clear that we will need regulation by product type and not institutional resolution and assessment powers, but subsidiary-level resolution and assessment powers. Charter type will no longer matter and regulatory coordination across functional subsidiaries will become paramount.
Nonetheless, the types of Chinese walls envisioned in Gramm-Leach-Bliley and traditional bank holding company regulation will remain of utmost importance and the main regulatory function will still be managing the "interconnectedness" of the firm as a whole, as well as the industry. Moreover, the academic understanding of economic importance of various levels of intermediation, insurance fund design, subsidy application and regulatory policy consistency will still apply. Policymakers ignore those universally understood principles at their peril.
