The ECB’s Stress Tests and our Banking Dis-Union: A case of gross institutional failure

stress tests imageLast Sunday the ECB published its quality assurance results, its stress tests of our systemic banks. It was, from where I am standing, a sad day.

Back in 2010, the European Banking Authority (EBA) disgraced itself with stress tests that passed with flying colours banks that would fail in short shrift in the months to come. What we have now done is to discredit the only proper institution that the Eurozone has: the ECB.

Am I being too harsh? I do not think so. Let us move, for a moment, to the other side of the Atlantic. What is it that gives the Fed and the FDIC power over banks? Is it the fact that the Fed-FDIC is the banks’  single supervisor? No. What gives the Fed-FDIC power over banks is the common knowledge that, when it assesses that a bank is insolvent, it has no serious qualms saying so. The reason, of course, is that it not only has powers of supervision (i.e. access to their books) but, crucially, powers of resolution and, if it so judges, the power to force mergers or to recapitalise the failing bank.

Suppose that, instead, the Fed-FDIC had, as the ECB does, only the power to scrutinise the banks’ books. Imagine now that, with one this power, the Fed-FDIC were to discover that some bank in Nevada or Missouri is in trouble. If the Fed-FDIC’s charter precluded it from doing anything else other than to announce the bank’s insolvency, its supervisory power would mean little. For if it were common knowledge that the fiscally stressed State of Nevada or Missouri would have to borrow from money-markets to pay for the depositors’ guaranteed deposits, as well as for any new capital the banks needed to be salvaged, the rest of the state’s banks would face a run, the states would see their borrowing costs skyrocket and, soon, a combined banking and fiscal crisis could be rummaging throughout the ‘dollar zone’. To put this crudely, the good people at the Fed would have no alternative than to keep their mouths shut, to conceal the bad news, to cover up for the bank’s problems and try to find some hush-hush way of bolstering its capitalisation.

This is precisely the sad state to which our, so-called, Banking Union has pushed the ECB’s supervisors into. As long as the ECB is not the sole authority on bank resolution, and as long funds for dealing with insolvent banks are to come (in the final analysis) from the fiscally stressed states, the death embrace between weak states and fragile banks will continue. And so will the credit crunch in the regions of the Eurozone where austerity is already taking its heavy toll on demand, having annihilated any urge to invest in productive activities – as opposed just to bargain basement unproductive assets.



  • You have produced not a shred of evidence that the stress tests were rigged, nor named banks who passed who are commonly inferred to be insolvent.

    • You are right. What I did do is to point out a major conflict of interest that was forced upon the ECB by the Banking Union’s structure. That should be enough. If you want more, I can happily provide it. For instance, while a drop in GDP was part of the scenaria used, deflation was not. Why? Because the ECB does not want to be seen to be discussing deflation, lest it triggers it. Another example of a toxic conflict of interest. Additionally, when the NPLs of the Greek (and other banks) are considered fully recoverable, one knows that these stress tests are not worth the paper they are written on. Of course, I am not in a position to check the books of every bank tested – and to offer evidence of ECB malfeasance on every count. However, I have looked at some. And what I have seen makes for distressed reading. For instance, did you know that the ECB approved as tier-1 capital a bond issued by an intermediary of a Greek bank which the said bank bought?

  • Any stress test like the one which the ECB has performed is a good thing because, that is for sure, it got the bankers’ attention. I can say that from personal experience: the Austrian bank from which I retired a couple of years ago was one of the six Austrian banks subject to the stress test. Whenever I met former colleagues during the last months, I noticed that the stress test was on the top of their minds. In short: they were nervous. And it is always a good thing when bankers are nervous. Nervous about making bad loans; nervous about buying the wrong securities; nervous about getting caught at doing untoward things.

    Having said that, it would be an illusion to think that all those banks which passed the test are ‘safe’. I looked up the figures of Deutsche Bank as of June 30, 2014. Their loans were only 23% of total assets. Financial assets of all kinds were almost 60% of their assets. Unlike loans where there are borrowers, behind financial assets there are counterparties and in the majority of cases they are not valued based on some notional amount or based on some tradeable value in a public market but, instead, based on complex formulae. If those formulae cease to be valid (like they did with LTCM back in 1998), the value of the assets is in the dark. More importantly, Deutsche’s consolidated equity was only 4% of total assets or a leverage of about 24:1. That’s a lot better than the 50:1 which it had a couple of years ago but it is still in the hedge fund category. Put differently, Deutsche is still a hedge fund with a small commercial bank attached to it. That’s a risk which cannot really be measured in terms of numbers.

    As far as I know, the stress test did not at all take into account the interrelationship between banks (i. e. they examined banks on their own merits). Anyone who is interested in what those interrelationships can cause is well advised to read up on the history of LTCM back in 1998.

    In summary, the stress test undoubtedly had a positive impact (at leat a short-term one) on the conduct of banks and bankers. However, anyone who concludes that the EZ financial system is on solid grounds is living in an illusion. The leverage of the large banks (mostly German and French) is far too high and they are far too reliant on ‘hot money’ for their refinancing. And structural issues have not been addressed at all. Many of the countries are totally overbanked leading to the situation where no bank on its own can make enough money without entering into undue risks. To me, the only answer is to increase the equity requirement based on notional balance sheet amounts (instead of only risk-weighted assets) which would lead to the necessary consolidations.

    • Thank you Mr.Kastner. Thanks for the facts, that make the situation understandable, even to the banking unexperienced among us. Also Yanis’ interview to RT gave us some facts as to how to value the stress test results and why. Here in Germany, the narrative on the banking issues, is more or less the same as with the hole economy. Underperforming economies with underperforming banks of the South are causing the crisis. “Our Deutsche Bank is the symbol of German reliability, professionality and competitiveness” they think, hopefully without saying it.

      This is the level of the public discussion going on in Germany, up to the level of responsible leaders and politicians. It is unimaginable for them, that the Deutsche Bank could be part of the problem. On the other side, they do not need evidence, they already have the prejudice, enforced by “quality journalism” and politicians, that the only the banks of the periphery are problematic and voulnerable. These stress tests will make them even more stubborn then they already are, in their prejudice. We can only change this with facts. Facts which at least the intellectuals, those who have the independent mind, but not necessarily the expert knowledge in banking, can understand. Only so we can over confront these fatal lies and myths, which we in Germany have very gladly believed.

  • We – Europeans – live in societies that are ruled by delusional people who are in the habit of constantly lying to themselves and their electorates, while trying to prove that their made-up arguments are real by citing other figments of imagination as evidence. Is it any wonder that the so called ‘stress test’ is just another one of these delusions?

  • Mr Varoufakis, I should say I am American, and also I respect your work, agree with the vast majority of your views, and see you as a man of integrity (as apposed to many in the EU hierarchy, which have near ZERO scruples, EXAMPLE: Herman Van Rompuy).

    I had questioned you before [many months ago] about Alpha bank (a Greek bank) before for selfish reasons. You gave me a blunt answer (which I am in gratitude for, for it’s honesty and most likely saved me losses). As I said at the time, I know you are not in the habit of doling out investment advice (nor are you paid salary for such service). HOWEVER, I was wondering if you had any thoughts positive or negative on Banco Santander. Good, bad, or perhaps a passing paragraph on how their balance sheet and capitalization was on a RELATIVE basis to other Euro Area banks???? Any thoughts you have as to Banco Santander IN PARTICULAR would be GREATLY APPRECIATED, for purely selfish reasons (of the investment nature) on my part.

    Warm regards.

  • All of these scenario based tests suffer from some serious flaws in their methodology. First of all, they describe a state of the world w/out clarifying the likelihood of this very state. Therefore, if a bank where to use some form of regime switching approach to integrate the baseline VaR and scenario based VaR calculations what is the expected VaR over all of the possible states of the world w/out having a measure of probability for each state? Therefore, they are rather meaningless from a practical application point of view. Secondly, most of them are derived from “expert judgement” and/or linear relationships between (assumed) jointly normally distributed risk factors. Another gross misrepresentation of reality. Finally, where is CVA in all of these? CVA was the single most important source of losses for banks during the Great Recession. Therefore, in my humble opinion internal bank practices (for those banks that have the IP and budgets) of measuring the tail losses using methodologies such as Extreme Value Theory coupled with non-linear inter-dependencies (e.g. skewed and assymetric copulas) are the best approach to identify potential tail catastrophic losses. Other good approaches include reverse stress testing and proper regime-switching (i.e. hidden Markov) models.

  • The EU passed a new regulation that the crisis is over. So the crisis must be over!

    • Exactly the same with “Zerohedge says there’s no austerity, so there isn’t any!”. Similarities are remarkable.

7 Trackbacks