In a recent interview with the Wall Street Journal, FinServ Chairman Hensarling said, “Certainly, we have to do a better job ring-fencing firewalling.” So far, big banks are mustering formidable firepower to beat back TBTF “cures” that would repeal OLA, set size limits or impose punitive capital charges (another idea Rep. Hensarling supports, by the way). These challenges are a real and present political danger. But, a far more potent critique of big banks is that use of insured deposits for non-bank activities gives these firms undue advantage that’s unfair to competitors and risky for taxpayers. Rep. Hensarling’s comment is no off-the-cuff remark – see the mandate to GAO in its TBTF study to look hard at this issue to see where change is needed. As the impossibility of executing the Volcker Rule becomes ever more apparent, Congress is readying Plan B: strict limits on the extent to which banks can use insured deposits for non-bank activities. Asked about this, many bank CEOs readily acknowledge that they shouldn’t use bank-derived advantages for non-bank activities. But, pushed to disentangle their business lines, most stop cold because their business model is fundamentally premised on efficient transfer of bank funding into non-bank operations. Making theory meet practice to craft a constructive response to industry critics is, I think, an arena in which reasoned discourse now could prevent drastic, unconstructive action later (or, maybe, pretty soon).

This issue isn’t new. When Congress first allowed bank holding companies to engage in a limited class of non-bank activities, it added Sections 23A and 23B to the Federal Reserve Act. These set a limit of ten percent of capital on the amount of funds that can be transferred from an insured depository to BHC affiliates and mandates that any such transfers be on an arm’s-length basis. This might seem tough, but it proved pretty porous, leading Congress in the 1999 Gramm-Leach-Bliley Act to tighten up the inter-affiliate transaction restrictions in concert with opening up lots more business lines to holding companies affiliated with insured depositories. Still unsatisfied, Dodd-Frank goes farther, putting some limits on the FRB’s ability to authorize transfers and giving the FDIC – long fearful of them – a greater say. Is this enough? As Rep. Hensarling states, not to those who, like him and his home-town Fed President, Richard Fisher, want stricter limits on funding transfers. The challenge isn’t so much the law or even regulatory acquiescence to bank pleadings for more transfers. Rather, it’s that so many non-bank activities are now conducted inside banks that the firewalls in practice have limited, if any, impact. In this, regulators are to blame at least as much as banks.

When first erected, BHCs were strictly confined to a very few non-banking activities and banks mostly just took insured deposit and made old-fashioned loans. Thus, Sections 23A and B stood between clearly-defined business lines and had their intended effect. But, as the OCC and some state regulators allowed more and more activities to count as the “business of banking,” the fundamental function of banks as intermediaries faded and their potency as financial powerhouses grew exponentially, fired in large part by the funding advantages derived from their bank charters. Now, open the lid on a bank and you see a dizzying array of securities, insurance, advisory, structuring and even commercial activities that would have made regulators faint only a decade or so ago.

A bit of this funding advantage might come from TBTF expectations – surely, much of it did before the crisis and subsequent enactment of OLA. But, even without any TBTF safety net, banks raise funds more cheaply than others for two reasons. First, regulated as they are, investors trust them, giving banks ready access to wholesale funding sources not as easily available to others, including even their own parent companies. Most fundamentally, though, the bank advantage lies in the fact that they are the storage facilities retail and business customers trust, not just because of FDIC coverage – critical to be sure – but also because banks have been the safe keeper for so long that they are category killers in this critical economic function.

If insured deposits weren’t of real use, would community banks exist? I doubt it. Would giant banks have thousands of branches and, now, be spending billions to beef up electronic deposit-taking systems. Again, I doubt it. And, are these charters worth it to others? You bet – just look at all the non-bank banks and the big firms – e.g., Walmart – that still want in. So, I think it’s an established fact that being a bank is still a government-granted benefit even if banks are so burdened by rules that it’s hard to remember this a lot of the time. Should this benefit be walled off as originally intended and now sought again?

I think the answer is yes in theory and sort-of in practice. The theory is straightforward: taking the public’s money serves a systemic purpose that warrants government-afforded benefits that should not be transferred to businesses that fail the public-purpose test. The practical challenge lies in figuring out where to build the ring-fences. Mr. Fisher merrily ducks this challenge when he calls for inter-affiliate transaction limits, but a meaningful proposal here must seriously and specifically take this on.

I think we need carefully to consider this and, for a change, big banks should lead the way. If they support the ban on proprietary trading in theory – most do – but hate the

Volcker Rule in practice – for sure – let’s define what should go where and ask Congress to set new rules for this complex road. Separating activities instead of seeking to ban them altogether won’t satisfy Mr. Volcker – he thinks banks too evil to be trusted with much other than operating ATMs. But, it’s a meaningful way not just to address risk-taking in systemic banks, but also to permit the useful economic functions these banks offer to continue with only one change – BHCs would have to find the funds to support these operations all by themselves.