It was a bit of a shock to me and many of our readers when the current trouble in JPMorgan's derivatives desk erupted. Not because the vulnerability to this type of problem in the desks of these big banks was not there - IRA's stress testing methodology has tracked off-balance sheet exposure as part of our CAMELS analysis regime for years -- but because I had not expected that this particular bank would be the one where this risk would first realize in the market place. In retrospect, that aspect of the letter S in the term CAMELS which stands for "sensitivity to market risk" is in fact uniformly distributed among the participants in derivatives market making and the susceptibility to a future beta event remains for any of them.
The question at this point is not whether the rest of the banks have this risk. The going forward questions are more appropriately, how should banks manage their susceptibility and vulnerability to this class of risk? How should insurers and markets price the risk-reward nature of such exposures? And in what direction should regulators aim the going forward definition of safe and sound practices?
IRA has commented a number of times in the past that we believe that risk and stress testing needs to be done in the context of benchmarking as opposed to the myopia of internally focused analysis. In this case, the need is even more acute given the fact that the banks (a) have trillions upon trillions of notional balances exposed and (b) bank counterparties view these derivatives exposures as material investments. So let us look at a catalog of banks so exposed. As is the case whenever it's an IRA analysis, we winnow from a census of all the active banks and look at the individual FDIC Certificate holders. In this particular illustration of systemic vulnerability, the unit institutions with assets over $10 billion - the Dodd-Frank stress testing and reporting threshold - that have derivatives operations that reported a fair value estimates of the traded portion of their exposures in the 1st Quarter of 2012 reporting cycle.
Table 1 - Over $10B Asset FDIC Certificate Holders, 1Q2012
NOT FOR TRADE
As can be seen, within this group, there are four large players followed by five medium sized players and then a collection of lesser - but still exposed - participants. All told, fifty three banks in this highly focused peering. Biggest of them in terms of the size of the notional derivatives book is JPMorgan Chase Bank N.A.
What is far more important to understand though is the context of the risk undertaken by JPMorgan versus this peer group. Was it extraordinary? Does analysis of it help us understand where the line of what is systemically unsafe might lie? For that we first turn to leverage. For this we look at the ratio of these institutions' traded derivatives to the fair value envelope of these instruments as reported in their CALL reports. The resulting number is an indicator of "the amount of scrambling that is likely to have to happen in the event of a glitch in the Matrix". The larger the total notional balance size combined with the leveraging factor help quantify the potential nightmare of each billion of realized loss. It's the kind of thing that makes a Tums and Xanax a food group for Chief Risk Officers; maybe for corporate treasurers too.
Table 2 - Derivative Desk Operating Leverage Multiplier Estimates, 1Q2012
OF TRADED BOOK
FAIR VALUE OF
NOT FOR TRADE
Notice that there are a variety of strategies with regards to derivatives that begin to be exposed by this relatively simple calculation. There are institutions that clearly pursue very conservative approaches to these instruments and others that make use of them more aggressively. That's not an unexpected result for anyone that has taken the time to understand that the pathways to operating a financial institution are not at all homogeneous, never have been.
Among the big four houses, there seem to be two schools of thought on this ratio, JPMorgan and Citibank electing to run their desks at one ratio and Bank America and Goldman Sachs operating at another one. It implies that a future beta event of similar magnitude as what beset JP Morgan would impact Citi similarly and the other two would have to scramble twice as hard. These are consequence management planning factors and are in fact most useful for costing how much to put into things like compliance oversight and risk taking authorization in the now so as to mitigate consequences if and when.
As one goes down the food chain the differences in strategy become broader straddling the middle ground of the big four with a few institutions electing higher leverage while most seek a more conservative path. Notable among these are the computed ratios of Well Fargo NA and Morgan Stanley NA which exhibit computational differences in the fair value estimates reported to the FDIC. Morgan Stanley basically says the balance sheet value is next to nothing. They may or may not be right but from a benchmarking standpoint, it is a real artifact in the numbers. A number of other smaller institutions also have this reporting approach.
Wells Fargo NA is notable because it pursues the most conservative leveraging approach even as the bank competes head on against its commercial banking competitors. In relative terms, the Wells Fargo desk would have to make a positioning error maybe five times the magnitude of what happened to JPMorgan to realize the same amount of turmoil. Interestingly, it indicates to someone like me to caution that Wells Fargo be the one most on guard against future complacency. Their competitors have incentives to be more hyper-diligent going forward and in the competitive space of derivative zero sum games, that means something.
Finally let's look at how much of the real balance sheet is affected by these derivatives. Using the assumption that the traded leverage ratio is a fair indication of leverage in the remainder of the derivatives book, we can perform a CALL report based apples-to-apples estimate of the book value of these instruments. The caveat is that this analytical assumption may or may not be true but confirming this would have to involve performing a proprietary review to increase the fidelity of the calibration. Still, the question of just how much a derivatives desk impacts the overall portfolio of a bank is an important piece of information to have.
Table 3 - Portfolio Analysis of Estimated De-Leveraged Derivatives Value vs. Total Assets
|NAME||TOTAL ASSETS, $K||ESTIMATED|
TOTAL ASSETS, %
So in context let's look again at JPMorgan Chase. It's definitely the biggest operation dwarfing every other. While the business was run on the higher end of the leverage spectrum, it was not the highest model in operation. In terms of tangible exposure to the balance sheet, it is or was the segment leader. The combination of the three factors is the signature of high susceptibility to event risk and high vulnerability to significant stress in the event that risk manifests. In the end it was a mistake to have ever believed that mass and reputaion were that much protection against this type of risk. It's a painful lesson to be sure and one that other banks should heed based purely an objective assessment of their own numbers in the context of these types of inconvenient truth benchmarks.
Like IRA's page on Facebook.
Reprinted with permission. Copyright 2012 Institutional Risk Analytics. All Rights Reserved.
The Morning Email helps you start your workday with everything you need to know: breaking news, entertainment and a dash of fun. Learn more