While most people scanned Friday's Board of Governors release on the Supervisory Capital Assessment Program, or SCAP, for clues to the stress tests, few noted the elements of a standardized approach in the making.
In particular, the last two pages of the release illustrate standardized schedules for reporting loan concentrations and loss rates that could make the procedure operational beyond the 19 biggest institutions. However, it would be premature to take the idea further without refinement. In my view three main areas exist where slight changes could make the procedure a meaningful and lasting part of regulatory reporting.
First, trading losses and counterparty risk evaluation can benefit not just from aggregation but also from the production of exposure schedules with accompanying risk estimations that — even if not made public — can help examiners piece together the "systemic" exposure of any institution in a reasonably timely way if a crisis arises. Furthermore, if constructed properly, comparisons of counterparty risk evaluations — where wide spreads between one bank and another's evaluation of a single counterparty could signal risk factors before they grow disproportionate — can benefit examiners.
Second, the asset reporting schedule merely aggregates existing call report data to the consolidated bank holding company level. The problem is that if a banking economist couldn't already do this with structure tables and call reports then they weren't worth their pay. Nonetheless, it is useful to have all the data in one place.
The SCAP schedule supplies a single marginal gain beyond the aggregation exercise by breaking first-lien mortgages into prime, Alt-A, and subprime categories. These distinctions fall short of providing clarity, however, because no regulatory classification exists for each category. Moreover, the subprime generalization lumps "good" subprime with "bad."
The operational definition is not hard to fix. The subprime classifications, however, require more attention. For instance, it would be desirable to know the proportions of option-ARM loans, no-doc loans, and LTV ratios and ARM spreads and resets in the subprime category. It would also be useful to start collecting data on reverse mortgages in this category; these are among the highest-growth products in today's marketplace.
Similarly, it would be useful to break credit cards, automobile loans and leases and student loans into prime and subprime segments. Strict definitions of each would help avoid the arbitraging of regulatory classifications so that a fast-growing "other" category should stimulate attention.
Though on-balance-sheet reporting is useful, however, in today's world off-balance-sheet exposures are paramount. Hence, the third area of improvement must lie in reporting not only what we understand as today's securitized loan exposure but also a watch for developing loan-sale and funding technologies. Rudimentary securitizations were captured on bank call reports beginning in 2004. Some securitizing banks were able to maintain on-balance-sheet asset exposures small enough to negate reporting, however, which was limited to "large" banks. Merely aggregating call report numbers to the consolidated holding company financial statements, therefore, would be of help. Still, the procedure lacks vigor in highlighting new forms of pseudo-securitization, like developments in structured investment vehicles and other infra-marginal means of moving loans, if not their risks, off-balance-sheet.
Banks also, therefore, need to report the amount of mortgage- and other asset-backed securities by tranche and rating level in order for investors to be able to gauge risk transfer via those mechanisms.
At the most rudimentary level, in an off-balance-sheet world regulators and investors need reporting on loan originations. Because loan originations have to go either on- or off-balance-sheet, a company's off-balance-sheet activity has to be its origination activity minus its on-balance-sheet activity. When a preponderance of that off-balance-sheet activity is going to places that can't be found, a substantial risk factor exists, worthy of investigation.
In closing, therefore, the SCAP schedules are a good start at regulating the new financial marketplace. They take elements of improved reporting and apply those elements to a dynamic capital market environment, borrowing from the interest rate risk stress tests applied to thrifts after the savings and loan crisis. It is a sensible approach that should benefit from the momentum built up by the exigencies of the credit crisis and the industry benefits conferred by the SCAP program. It is the first attempt at regulatory changes for today's financial marketplace, and it should not get lost in the policy shuffle.
