The holes of the BIS rule book

At Sober Look we always appreciate your e-mail. It's part of what makes Sober Look what it is - everyone's view counts. An e-mail came in at 2:45 this morning that was nearly 20 pages long from Mr. Anonymous. Thanks for the e-mail Mr. Anonymous. It was difficult to connect all the various stories of the e-mail, but the beginning was a classic:

"with your luxury of anonymity...rebut this:

"The Hidden Beginning" by Bruce Wiseman: On April 2, 2009, control of the planet’s banks was turned over to the secret decisions of eleven men—board members of a Swiss organization with a troubling Nazi past..."

Hmmm. Only 11 men? No women? Seems Bruce Wiseman is talking about the Bank for International Settlements or BIS. The BIS board members are in fact the central bankers from various nations. With regard to the 11 men, seems we have a few more these days (here is a quote from BIS):

"The Basel Committee on Banking Supervision decided to broaden its membership and to invite as new members representatives from the G20 countries that are not currently in the Basel Committee. These are Argentina, Indonesia, Saudi Arabia, South Africa and Turkey. In addition, Hong Kong SAR and Singapore have also been invited to become members. The Basel Committee's governing body will likewise be expanded to include central bank governors and heads of supervision from these new member organisations.

With its current expanded membership, the Committee is now comprised of representatives from Argentina, Australia, Belgium, Brazil, Canada, China, France, Germany, Hong Kong SAR, India, Indonesia, Italy, Japan, Korea, Luxembourg, Mexico, the Netherlands, Russia, Saudi Arabia, Singapore, South Africa, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States.

The newly expanded membership will enhance the Committee's ability to carry out its core mission to strengthen global supervisory practices and standards. It will also help to more effectively implement the necessary reforms of the international financial system. "

The mission is in fact "to strengthen global supervisory practices and standards". Among other things, these folks are known for what's called the Basle Accord. Basle set the standard for determining the amount of capital banks should hold. The local central banks would make adjustments for financial institutions within their jurisdictions, but in general followed the capital rules of the Basle Accord.

Driven by academics, the organization strives to bring a near scientific sophistication to bank capital rules. Their big achievement was to assign capital requirement for "trading book" assets (as opposed to banking books). The formula was a multiplier times the value at risk (at 99% confidence interval and 10-day holding period). Sexy hah? Banks spent hundreds of millions implementing and complying with this - trying to bring all of their trading positions into a single VAR number every day. Then they would multiply this number by the multiplier (which usually was between 3 and 5 as assigned by their local regulator) to get their capital requirement. All nice and scientific.

The "banking book" was however another matter. It's harder to "model" a bunch of corporate loans. So under the first accord, the formula was simple. You have a corporate loan on the books, you compute 8% of the face value, and that becomes your minimum capital requirement. That was it until the newly minted Basle-II accord that bases banking book capital requirements on ratings either internal or better yet rating agency ratings. Isn't that special? Basle-II is still being implemented by a bunch of banks.

But most banks were still under the original Basle accord as we were entering the crisis. The bulk of the capital requirements for banks was coming from the banking book (loans originated by the bank). Let's look at an example: as a bank you lend to a corporation at LIBOR + 1% - for a strong corporate credit. You borrow at LIBOR in the interbank market. So your income is 1% of the loan face, and according to Basle, you have to put up a minimum of 8% in capital. Most banks wanted to put up more than the minimum to keep their regulators happy and the stock analysts shouting "buy!". So most had some 10% or more in capital. But that's a bit of a problem. Your return on capital is now 1% (income)/10%(capital) = 10%. 10% return on equity is way too low. Shareholders wanted more, way more.

The banks needed a loophole and one was readily available from BIS. The capital requirement drops dramatically to a fraction of the 8% if
1. Your corporate loan is unfunded - that is it's an unfunded commitment or a guarantee.
2. The commitment (above) is under a year in length.

So how do you get a 7-year loan to satisfy the requirements above? The answer is what's called a "Commercial Paper (CP) Conduit". You create a company and gift it to a charity. That company will do the lending instead of the bank. It will fund the lending by issuing commercial paper, short-term obligations (under a year). But why would anybody buy this commercial paper? Because the bank would provide a guarantee (that satisfies the two requirements above).

Based on that guarantee, the rating agencies would give the commercial paper a high rating, making it marketable to money market funds. In return for the guarantee, the bank would get to keep most of the spread, the difference between the loan interest and the commercial paper cost (with a sliver going to the charity who owns the CP conduit). Now the bank makes the same income, but on a loan that is not on it's balance sheet (the loan is on the balance sheet of the CP conduit owned by a charity). The bank's only capital requirement is based on that guarantee to the conduit, which is under a year and unfunded. Capital requirements are now closer to 2% rather than 8%, catapulting the return in our example to 1%(spread)/2%(capital) = 50%.

Now we are talking. Let the charity do the lending, the bank provides a guarantee and gets massive returns. The shareholders are happy, the charity is happy, and the regulators are clueless. This was called Regulatory Capital Arbitrage. Clever hah? While BIS and the regulators focused on VAR for the trading book, the banks ran a shadow bank on the side and pumped up returns. All perfectly legal, all within the BIS rules.

Of course it wasn't just the corporate loans that went into these CP conduits (there is only so much corporate lending a single bank can do). It was also the senior structured credit bonds, securitized by mortgages (mostly subprime mortgages with a nice yield) and other consumer loans. The same trick: CP conduit buys the bonds, issues commercial paper, bank provides a guarantee, gets low capital requirement. The tighter the spreads became the larger the size and the number of the CP conduits. Hundreds of billions.

Then in 07 the music stopped. As subprime default rates picked up in early 07 (a story for another day), the commercial paper (CP) buyers got nervous. Since CP is a short term loan and needs to be rolled every 1-6 months, the buyers all of a sudden said... not this time. With no commercial paper to finance the assets, the guarantees from banks kicked in, forcing banks to fund all the assets in the conduits directly from their balance sheets. But remember, based on BIS rules, banks only had a couple of percent in capital against these assets (50 x leverage). And now the banks owned these assets directly, as the values kept deteriorating. Most of these were (or became) highly illiquid assets. We all know what happened next.

Banks followed BIS capital recipe precisely and found themselves massively under-capitalized (or overlevered). It was all legal and by the book, the BIS book. Thus in many ways BIS (together with a bunch of factors we plan to discuss later) may be responsible for the financial crisis. So Mr. Anonymous, back to your e-mail. The control of planet's banks was not turned over to BIS on April 2, 2009. Unfortunately the control got turned over to them much earlier (back in the 90s), as the national regulators increasingly relied on Basle to keep their banks properly capitalized. Our only hope is that going forward, bank regulators use some common sense and ask fundamental and practical questions, rather than blindly putting their faith in a rule book devised by a bunch of academically minded bureaucrats in Switzerland.

Update from Mr. Anonymous:

"Perhaps you could clarify for your readers that I was the messenger not the author. Although the temptation for ad hominem attack was mistakenly directed at me I appreciate the time you took to write your very instructional clarification of Wisemans piece."