Tuesday, November 2, 2010

On Forcing all broker-dealers to go private (BAC; C; GS; JPM)

Back in March '09 We expressed similar thoughts regarding Goldman Sachs.*
From Felix Salmon at Reuters:

I’m back! And I couldn’t be happier with the fantastic set of guest blogs from Justin and Barbara — wonderful stuff. If you haven’t read it, for instance, check out Barbara’s post on rules-based vs principles-based regulation, especially as it applies to the Volcker Rule. Volcker himself advocates a principles-based approach, contra Michael Lewis, who wants some very tough rules:
Here’s a simple, straightforward way… to construe the Dodd-Frank language, and it would reform Wall Street in a single stroke: to ban any sort of position-taking at the giant publicly owned banks.
Our crisis was not drastic enough to enable legislation that ambitious, but in theory I like this idea. Basically, it forces all broker-dealers to be private rather than public companies. That was the case before Bear Stearns went public in 1985, so it’s clearly entirely possible. And Lewis points to Citadel as a good example of a private broker-dealer dealing very successfully in the much larger and faster markets of today.

Broker-dealers as a set might well get smaller if such a rule were enforced, but that’s a feature, not a bug. In fact, if broker-dealers don’t shrink at all, then the Volcker Rule has clearly achieved nothing at all.

More generally, I suspect that a lot of people who blame Gramm-Leach-Bliley (the repeal of Glass-Steagal) for the financial crisis should really be blaming the broker-dealers going public instead. After all, Bear Stearns and Lehman Brothers and Merrill Lynch were both entirely Glass-Steagal compliant, as, for that matter, were Fannie and Freddie and AIG. The problem wasn’t that they were merged with commercial banks; the problem was that they had far more leverage than any private partnership would ever be comfortable with....MORE
*David Viniar, CFO of Goldman Sachs Blows Smoke at Journalists on AIG
Goldman held their conference call with the journos and I am sure there will be commentary far more erudite than anything I could muster, so I will focus on just one small piece of the call. MarketBeat's David Gaffen has the link to Deal Journal's live-blog and says:
...David Viniar, CFO of Goldman Sachs, basically suggested that since nobody knew that AIG was a house of cards, nobody had any reason to suspect anything. “AIG was a triple-A rated company, one of the largest and considered one of the most sophisticated in the world,” he said. And in a response to a question on how Goldman allowed its exposure to AIG to get as large as it did, Mr. Viniar describes positions made in 2006 and early 2007 as if it was a different age, a more innocent time, when magical dwarves ruled the land, before the ring of power had been forged. “It was a very long time ago,” Mr. Viniar says. “AIG at the time was one of the largest, strongest companies in the entire world, and they were very sophisticated, or appeared to be, a very sophisticated counterparty.” The firm later scaled back trades — by the end of 2007, according to Mr. Viniar....
My question is, "If Goldman Sachs were still a partnership, would they have entered into these transactions in the same size?"

The answer, of course, is no.

If partners equity were at risk, there is no way that they would have depended on ratings agencies to ascertain the strength of their counterparty.

Junior partners would be expected to run honey traps on AIG employees.

Lower level employees would hone their dumpster-diving skills.

Whatever it takes to gain competitive intelligence and safeguard the partnership's capital.

See also: "The optimal design of Ponzi schemes in finite economies"
Reading Mr. Viniar's words, I am reminded of his statement on market moves in August, 2007:
“We were seeing things that were 25-standard deviation moves, several days in a row”
Several folks, when they finally quit laughing, pointed out how blatently Mr. V was spinning.
Most however underestimated how infrequent 25SD events are, the most common guess being once in 100,000 years. Tee hee.

In a snappy little eight page paper "How Unlucky is 25 Sigma" we see that at 7 Sigma the odds are...