Thursday, August 2, 2012

Commercial banks or hedge funds?

To those who are not familiar with financial statements, let me briefly explain what "leverage" is. Leverage is the relationship between total liabilities (rough translation: total debt) and the equity of a bank. It is typically expressed as the ratio of "x to 1", whereby "x" stands for total liabilities and "1" stands for equity. Obviously, the higher the "x", the more risk a bank carries.

The frequently mentioned minimum capital ratios according to Basel-2 are something different. First, they are expressed as "percentage of". For example, a 9% capital ratio means that the bank must maintain capital (a more narrow definition of equity) of at least 9% of risk assets.

The trick is that Basel-2 excludes certain risks from "risk assets". For example: Basel-2 deems sovereign debt to be risk-free. Thus, no reserves need to be held against it and it does not count as risk assets in the minimum capital calculation.

As an extreme example, a bank might have a Basel-2 capital ratio of 15% (far above the minimum required) but it might have, at the same time, a leverage of 20 to 1 (which is very high as explained below). How could that be? It would happen if the bank held a very large amount of sovereign debt.

A traditional banker would disregard the sophisticated Basel-2 definitions and simply ask "What's the bank's leverage?" The higher the leverage, the greater the risk. Every financial claim on a bank's asset side represents risk, even if it is the sovereign debt of a AAA-rated country. Why? Because assets need to be funded; the higher the leverage, the greater the funding risk.

There are no regulations as to how high a bank's leverage can be. A general rule is that it should not be too much above 10 to 1. Another general rule is that a leverage of 20 to 1 or more should be reserved for hedge funds.

Below is the leverage calculation of two banks each from the US, France and Germany.

JP Morgan:                 11 to 1
Citigroup:                     9 to 1

BNP:                          22 to 1
Credit Agricole:           21 to 1

Deutsche Bank:          39 to 1
Commerzbank:           26 to 1

A sample of two per country isn't necessarily a good basis for a theory but the two banks chosen do indeed reflect very much of the banks' structures in these countries.

The European banks are at least twice as much leveraged as the US banks. All of them are clearly in the category where one would expect hedge funds to start but Deutsche Bank would even be a trend-setter among hedge funds.

These kinds of hedge funds are the institutions which European tax payers (including Greek tax payers) have been bailing out over the last 2-3 years (and if Mr. Draghi get his way, they will even do much, much more bailing out of hedge funds).


  1. "The trick is that Basel-2 excludes certain risks from "risk assets". For example: Basel-2 deems sovereign debt to be risk-free [...]"

    Just a minor point: that is not entirely correct. Sovereign debt is automatically considered risk-free ONLY if it is issued by a member of the EU and in national currency (eg: Greece->EUR, but Sweden->SKR). Beyond that, credit ratings apply. Sovereign debt would then have to have a credit rating of at least AA (if memory serves right) to be considered risk-free.

    So whenever Greece issues a bond in USD, as it has happened, because of her awful rating, that debt would actually incur a severe penalty, namely a risk-weight of 150% (which is higher than the default risk weight of 100% for unrated debt, and on par with the risk weight for PE and hedge funds).

  2. Very interesting, it would also be nice to know where the risks of the DE & FR banks originated.

    Reading various sub-prime port-mortems it seems to me that Deutsche in particular was very naive, kids in a toy-shop, or am I being far too naive.

    Also I'd be interested in seeing a couple of banks from the City Casino - Barclays and HSBC perhaps.

    Off Topic Question

    As I understand LIBOR - the banks call the BBA in the morning to say "If we wanted to borrow some money, this is the interest rate we think we would be charged by other banks". The BBA discards the top & bottom outliers and computes an average the rest.

    Q1 : is the average a simple average, or is it weighted, if latter then what is basis of the weighting

    Q2 : wouldn't it make more sense to ask the banks to say at what rate they would charge banks A, B, ... J, Where A-J would be a random selection of banks determined each morning by the BBA, or better yet an independent authority - like a University Maths faculty ;)

    I don't pretend this would prevent collusion, fraud, etc - but as an engineer it would seem to a more sensible way of measuring the "mood of markets" - I realize this would mean more computations but ...


    1. 23 to 1 for Barclay's and, surprisingly, 14 to 1 for HSBC.

      When Germany went through its economic crisis in the early 2000s, I was working there. The large German banks had basically written of the Mittelstand as a target customer group because they considered it too risky. Deutsche Bank once had the nerve to state publicly "it cannot be the role of Deutsche Bank to finance the German Mittelstand!"

      One of the reasons why they took this attitude was that they had so many alternatives for better investments, such as highly rated sovereign bonds of EZ-countries which were deemed to represent no risk and which required no reserves...

      I have been dealing with rates like LIBOR or EURIBOR throughout my career. What we knew about it was that those rates were set every day at 11 am (I believe) by the largest banks in the interbank market. So it was stipulated in loan agreements. The thought that those rates could be manipulated never crossed anybody's mind. It was simply taken for granted that manipulation of such rates was technically impossible. Also, mind you that when doing business with customers in the real economy, one basis point up or down was irrelevant. Rates were typically set to the closest round 10 basis points (like 5,40% or 5,30%). In extreme cases the closest round 5 basis points. It just didn't make a lot of sense to spend phone expenses on haggling with someone over 1 basis point up or down when you are talking about principal in amounts of 5-10 million.

      When, however, a bank's treasury needs to refinance its entire loan portolio, one or two basis points make a big difference, particularly over time. My understanding is that the manipulation was about individual basis points and not about several basis points, not even to mention something like 0,10%. With customers in the real economy, that would have fallen under the category of rounding differences. With huge tickets, you are talking about real money for individual traders in a bank's treasury.

    2. Re Deutsche - So the Mittelstand was too risky, but not sub-prime mortgages they the bundled into derivatives to sell to the naive German banks who were supporting the Mittelstand. See next para re HSBC.

      The HSBC number doesn't surprise me entirely. It indicates there's a strong residual Asian ethos. Chinese people love to gamble and many are very successful doing so. Hence I assume they know how to assess risk.

      Re LIBOR - I appreciate that any advantage the banks got was on tiny fractions, and I'm not entirely convinced that their fiddling had any impact on the real economy.

      But it's the whole basis of LIBOR which I think is scandalous. Its seems foolish to me to ask someone what rate they think they can borrow at - its an open invitation to LIE MORE. And doubly foolish to have the London Bankers Social Club administering the whole thing.

      Why not ask a random selection of banks how much they would charge to lend money to random selection of different banks and get an independent body to do the sums.

      It reminds me of Yes Ministers Sir Malcolm Glazebrook - "Well they are all jolly good chaps. y'know.'

      I recently read Suzanne McGee's Chasing Goldman Sachs, very interesting, - my take out was : banking became a sport, where it was a race against Goldman's. McGee starts the race on Mayday 1975 when the SEC rules bought an end to fixed commissions on share trading.

      Thanks Klaus

    3. Regarding Deutsche, I presume you are referring to IKB. Yes, that was a scandal, indeed. A scandal which was never really given any public attention.

      IKB, to me, was the prototype of a solid bank. One of the best bank cultures I have met in my career. Their niche was long-term financing of investments in the real economy. Mostly Mittelstand, but not only. They were partners for their customers, indeed. One of the reasons they could be "partners" (i. e. not run away when the first raindrops fall) was that they always secured their loans well. That was accepted by the market because the market knew that this was a fair price to pay to have a solid partner. Actually, not a bad philosophy for lending...

      Customers could totally rely on IKB. When a customer got into financial trouble, one could rely on IKB as the solid force in a financial restructuring. I could go on and on.

      IKB was probably the best proof that, in a bank, a handful of people can do a lot of damage. I had many friends at IKB and we did a lot of business jointly. I would say that 99% (or more) of the bankers there did excactly the kind of business which bankers should do and exactly in the right way.

      Then they got a top management whose origins were in investment banking. Secured long-term financing for the real economy was not a sufficient challenge for them, also because it was not as profitable as some other things.

      A minute part of total staff got involved in SIV-structures. Since they had the liquidity to be buyers of assets, they were loved by those who were sellers/placers of assets. As we learned later, NY investment bankers were so taken by that that they asked "who are those guys in Düsseldorf?"

      Deutsche stuffed IKB with about any bad asset it had available and they were a major funder of IKB. Thus, Deutsche had a totally clear picture of IKB's risk portfolio.

      One of my friends told me the following: on a Friday afternoon, Deutsche informed IKB that they would cut their interbank funding for IKB the following Monday. Being the noble and responsible institution which Deutsche is, they had the kindness to also inform the banking supervision of that. Of course, that meant that the following Monday IKB was going to be out of business. Which it was!

      I have studied the events surrounding IKB in detail. My bottom line is that IKB's management were not crooks. They were simply dumb and naive. The prototypes of useful idiots.

      During my time in American banking, the saying was that the financial world consists of stuffers and stuffees. The stuffers originate assets of all kinds. The stuffees are those who buy them. The stuffers collect enormous fees and carry no risk. The stuffees got often invited into exclusive bank dining rooms as appreciation for their cooperation, and they loved that. Also, the stuffees could build up large volumes of assets without any hard work, really, and they could collect excellent interest margins. Those loans were very profitable until they turned out to be losses...

      Re impact of LIBOR-fiddling on the real economy? Well, start with the premise that whatever banks do even with one another, at the end of the day it is the customers in the real economy who pay the bill. If governments implement a transaction tax, it will not be the shareholdes who pay for it; the customers instead. When you have contractual commitments for a variable rate, you need to have a reference rate. No system will ever guarantee honesty if its users are crooks. If you fiddle the rates upwards, somewhere and somehow, at the end of the day, real customers will pay for it.