Chastened by his “shellacking,” President Obama is heading off to join fellow heads of state at the G-20, packing a briefcase of communiqués concluded in closed-door negotiations this week. The biggest unresolved issue to be smoothed over with neutral language after the meetings next week is currency policy (read, China). Decisions on financial-regulatory matters are either done – the Basel III capital rules – or at an impasse. Deadlocked issues include what to do about cross-border orderly resolution for systemically-important financial institutions (SIFIs). Reflecting this, each of the G-20 nations, with the U.S. leading the way, is retreating to a barbed-wire approach to cross-border resolution: “subsidiarization.”

What could a word like that mean? Subsidiarization is among the most ungainly coming out of the financial crisis, joining an already crowded lexicon – “macroprudential” anyone – of words crafted by supervisors with a better ear for what sounds good in a doctoral dissertation than English. What subsidiarization means in practice is breaking up complex financial institutions, including branches that cross borders – into distinct subsidiaries. This is often called “ring-fencing,” a term that makes clear that the goal of subsidiarization is to define robust boundaries between different corporate operations to keep the flock on one side and the wolves on the other.

This sounds straightforward – tell banks to pull themselves apart so that systemic entities are clearly identified and, where possible, broken into constituent parts to facilitate orderly resolution. Insured depositories would then go into one box – or, more precisely, into lots of boxes depending on in which nation each bank operation is housed. Non-bank components of a SIFI would similarly be stripped down into constituent pieces based on national jurisdiction, legal charter, line of business and similar criteria. In practice, however, pulling SIFIs into separate pieces is very, very complicated and not just because SIFIs like themselves pretty much as is.

The debate yesterday at an FDIC/FRB-sponsored roundtable showed clearly just how much supervisors want ring-fencing and how hard most in the industry think it will be to meet them even half-way. To be sure, some large, global banks are now more or less structured as holding companies with each of their operations in separate subs. This was cited by some as evidence that subsidiarization isn’t all that hard (assuming we can all learn to pronounce it). Left largely unsaid, though, is the fact that EU banks are sub-centric because European law largely required this before the relatively recent integration of the EU. Even now, most of the rules governing EU-wide financial activities force them into subsidiaries defined by business line and national boundaries.

This artifact of European history makes EU banking far less efficient than U.S. financial services. In fact, the sub-structure of EU banking is so inefficient that one principal goal of the EU in this area is to create “passports” and other criteria so that financial institutions can operate with branched networks, not separate corporate entities. Not coincidentally, supervisors have pushed for this structure in part because it also facilitates safety and soundness across the EU. Nation-by-nation regulatory schemes and resolution backstops (such as they are in the EU) create trapped pools of capital and liquidity that have difficulty crossing borders to strengthen weak links in a financial-company chain, especially when home and host country supervisors circle their own wagons and bar fund transfers.

We know that financial institutions have long rationalized risky structures on “efficiency” grounds and this is to some degree true too in the subsidiarization struggle. It’s one thing to have a complex structure based on persuasive business rationale – it’s quite another to have so complicated a company with so many cross-cutting entities with exposures to uncounted counterparties that no one knows who’s in charge of what where. “Efficiency” is also no justification for using resources in firms protected by deposit insurance or other taxpayer-supported backstops to subsidize other holding-company operations that compete head-on with non-bank companies without a comparable safety net.

But to agree that “efficiency” arguments can be misused is not to say that they are wrong-headed. Efficiency is a key tenet of effective markets that meet borrower and macroeconomic need. If supervisors simply force SIFIs into silos, they will undermine much of the real benefit proven by market demand for complex, multi-purpose, cross-border financial companies. To balance two vital policy goals – not just safety and soundness, but also efficient financial markets – regulators will need to improve their own operations even as they force banks to do the same.