Congress -- are generally thought to implement the policy of separating banking and commerce, the opposite is true. In fact it is not overstating the case to say that these bills eviscerate the separation principle.

For those of us who have argued for years that the only worthwhile banking "reform" legislation was a single sentence -- "the Glass-Steagall Act and the Bank Holding Company Act are hereby repealed" -- this is a delicious irony.

Fundamentally, the separation principle rests on a theory of economic harm, some way in which the combination of banks and commercial firms is likely to cause harm to the bank, the economy, or both. Since we do not have a general principle against combinations -- conglomerate mergers and acquisitions may not be universally salutary, but they are not illegal -- this theory must rest on the idea that some kind of specific harm will come from permitting banks, in particular, to combine with commercial firms.

This supposed harm has been catalogued repeatedly for Congress by advisers and commentators as eminent as Paul Volcker and Henry Kaufman. In testimony to congressional committees in recent years, Mr. Volcker identified three dangers of permitting banks to affiliate with commercial firms: The bank with a commercial affiliate will lend preferentially to the affiliate, whether willingly or under duress from a commercial parent; the bank with a commercial affiliate will not lend to competitors of its commercial affiliate; and if the bank's commercial affiliates get into financial difficulty, the bank's resources will be marshaled to bail them out.

I happen to believe that these hypotheticals (in addition to positing violations of the banking laws) are wholly fanciful, but for purposes of this analysis we can assume that they are valid.

What do the proponents of separation mean when they say that banking should be separated from commerce? Banking, after all, could be considered a form of commerce. However, we understand that the phrase is not to be taken literally but to describe a separation between the users of credit and the government-insured suppliers of credit.

The statement really implies that it would not be good policy to let the users of credit control the suppliers, or vice versa, because certain bad things might occur. If we go back to Paul Volcker's three examples, they all describe the bad things that might occur if, say, a bank controlled or was controlled by an automaker.

Turning, then, to HR 10 and S 900, the distinguishing feature of both is that they permit banks to be affiliated with securities firms and insurance companies. Though it was hotly disputed whether this affiliation should occur through holding companies or bank subsidiaries, the idea that such affiliations should be permitted was not controversial.

Now in what sense would it be true to say that this is consistent with the principle of separating banking and commerce -- which we understand to mean the separation of the suppliers from the users of credit?

Are securities firms users of credit? Certainly. Are they users of credit in the same sense that, say, an automaker is a user of credit? Again, certainly. The fact that securities firms, as financial intermediaries, are also suppliers of credit is not relevant to the analysis. In the same sense, automakers are also financial intermediaries and suppliers of credit through their subsidiary finance companies.

Just to be sure we have this right, then, let's test it against Mr. Volcker's three hypotheticals. Could a bank that is controlled by or under common control with a securities firm be required to lend preferentially to that firm? Yes. Or refuse to lend to competitors of that firm? Of course. Or if the securities firm got into financial difficulties, could its affiliated bank, as Mr. Volcker suggests, be importuned to make funds available to bail it out? Certainly.

And would the same thing be true in each case if it was an auto company that was affiliated with the bank? If you believe the Volcker hypotheticals, the answer has got to be yes.

So is there any difference, from the standpoint of the harms that the separation of banking and commerce is intended to prevent, between a bank's affiliating with a securities firm and the same bank's affiliating with an automaker? It seems obvious that the answer is no.

If this is true, in what sense can HR 10 or S 900 be said to preserve the separation of banking and commerce? The concept that seems to underlie these bills is that banks may control or be controlled by companies that are engaged in financial activities, some of which are specified and others of which are draft choices to be named in the future by the Fed. This is an apparently reasonable place to try to draw the line -- if you believe for some reason that a line must be drawn -- but no one should be under any illusion that this idea implements the principle of separating banking and commerce.

Nor should anyone believe that the Fed has been given a standard in either bill that can be readily enforced. As the auto finance company example illustrates, there is simply no principled basis for deciding what is and what is not a financial activity. In Regulation K, for example, the Fed declares that computer consulting and software development are financial activities, demonstrating both the squishiness of the financial-activity standard and offering perhaps a theoretical basis for Microsoft to acquire a bank.

It is time for Congress to recognize that with the abandonment of the separation of banking and commerce there is no longer any basis in policy or principle for the continued existence of the Bank Holding Company Act. Instead of trying to patch it up this tattered and tired old idea, it should be repealed along with Glass-Steagall.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.