The new Treasury blueprint for reforming financial regulation is not really a blueprint at all. (The full document is here; or see a Treasury summary of it here.) It says some sensible things and has some good ideas, but for the most part it is an agenda for discussion rather than a detailed plan. Given that the Treasury has been working on this thing at least since March 2007, it is surprisingly thin.
Moreover, it is concerned exclusively with the structure of the regulatory system. I think that getting this right is more important than Paul Krugman does—he calls this the Dilbert strategy—but Paul is surely right to complain that a better structure will get you only so far. It is a question of form and content. What the rules say matters more than which regulators are responsible for enforcing them, and the so-called blueprint does not go into that.
For instance, the document calls in the short term for the Federal Reserve to “continue to write regulations implementing national mortgage lending laws”. It also “recommends clarification and enhancement of the enforcement authority over these laws”. Fine. But what should these laws actually say? We are told only that they “should ensure adequate consumer protections”. No doubt; so far as I am aware, nobody is calling for inadequate protections. The question is, what does “adequate” demand. Up to now, the Fed has presumably judged the existing standards—embodied mainly in the Truth in Lending Act—to be all right. (When he was Fed chairman, Alan Greenspan applauded the explosion of subprime mortgage lending, saying it let people buy a home who would otherwise have been unable to.)
In the long term, the report envisages a structure based on what it calls an objective-based approach. This means dividing up the regulatory workload by broad tasks, rather by types of institutions covered. Today’s approach segregates banking, insurance, securities and futures under different (overlapping and cross-cutting) systems of regulation—an idea that last made sense decades ago. The Treasury urges instead a three-part division of responsibilities. There would be a “market stability regulator”—the Fed—whose job would be to gather information and monitor risks across the whole financial system. There would also be a single “prudential regulator”, concerned with capital adequacy, investment limits, and risk management—with a remit confined to firms with explicit federal guarantees. Finally there would be a “business conduct regulator”, operating across all types of financial firms (and absorbing the SEC), to protect consumers and investors.
This basic structure makes much more sense than the present chaotic array of industry-focused regulators. But one big gap—even at this level of generality—is obvious. On my reading of this document, the Fed is seen mainly as a system-wide information gatherer, not a rule writer or rule enforcer. True, it would have “the responsibility and authority to…take corrective actions when necessary”. But what does that mean? I think it means, provide liquidity at times of stress—which it does already. The document does not explicitly entertain the idea that the Fed would write and enforce new rules to make stability easier to achieve. The prudential regulator would do that, you might argue--but only for firms with federal guarantees. That narrow remit would have excluded Bear Stearns, for instance. Investment banks and hedge funds will not be covered. And the business conduct regulator does not fill this gap either: it is concerned with consumer and investor protection, not financial safety and soundness.
It seems to me obvious that prudential regulation ought to extend beyond firms with “explicit government guarantees”. At the very least, delete “explicit”. We are witnessing right now how the collapse of firms without explicit guarantees may nonetheless pose a threat to the integrity of the whole financial system. Evidently, it is a threat that the Fed and the Treasury have recognized—and that is why the umbrella of implicit guarantees continues to expand. There must be a regulatory quid pro quo for that, just as for the explicit kind.