El Confidencial 26/7/14
“Stress tests are like Cuban universities, everyone passes, but the title is worth nothing.”
If anything has been shown in the recent episodes of negative surprises from some European banks – from Portugal to France or Germany – is that European banks have not yet solved their problems. We confuse the important exercise of transparency and improvement carried out since 2012 with a magic solution to a problem created by a decade of excess. Is impossible to assume that banks have cleaned up their balance sheets when non-performing loans reach EUR 932 billion across Europe, 7.6% of total loans in the eurozone, according to Price Waterhouse Coopers .
When we talk about the bank stress tests (“stress tests”), many analysts regard them as the definitive marks, not as what they really are: a dynamic analysis of constantly changing circumstances . And, of course, these tests are not infallible, as we have seen so many times (Dexia, Savings banks, Cypriot banks, etc..).
WHAT IS A ‘STRESS TEST’?
It is an analysis that uses a common methodology for all countries, in which the impact on the capital ratio of a bank of various risk events is analyzed .
The general public tends to think of banks as “entities that collect deposits and make a lot of money,” and this is wrong. Because we tend to look at the P&L (accounted profit and losses), and not the cash flow and balance sheet.
A bank, for every Euro received from deposits usually borrows up to 25 times. That deposit is actually a loan, it is not sleeping in a safe.
For every Euro that the entity gives as a loan, banking rules allow the use of more or less capital depending on the risk that is assumed for the operation. If the bank lends to a very safe company with a low probability of default, the percentage of capital required for the loan is very low. The rest is debt.
These loans, if they work, generate a profit, and during the life of such loan the bank generates a margin between the cost of money and the interest rate charged… if the bank gets the principal back. If not, the balance sheet will deteriorate rapidly.
Whhen things go wrong, such “capital” shrinks very quickly. This is why people do not understand how in 2007 a bank could have reasonable solvency and liquidity ratios and in 2008 be on the verge of bankruptcy. A citizen does not see how fast the core capital can disappear when a large percentage of loans become risky (non-performing), which means that there is a high degree of probability that the borrower will not be able to pay the interests and principal. This rapid decline in the middle of a recession can leave the bank without resources.
As such, stress tests aim to analyze whether in a drastic change in the economic environment, banks would retain the 10-11% of capital that they have on average today.
WHAT DOES A ‘STRESS TEST’ MEASURE?
The stress tests aim for two objectives. Firstly, analyze the impact on banks’ fragile financial structures of events like a recession, losses on sovereign bond portfolios, aggressive currency depreciation, etc.. Furthermore, the test stress tries to be righteous enough to not make an unnecessarily negative exercise that endangers the public trust in the institutions.
WHAT DO BANKS FACE IN THIS ROUND OF TESTING?
Many large banks are currently generating returns of around 4% (return on equity), far below the typical target levels of around 15%.
Research by EY suggests that banks will find it extremely challenging to achieve this kind of RoE uplift. Cost reductions of around 35% or revenue growth of more than 20% might be required just to achieve their average cost of equity (10%). Should banks wish to reach 15% RoE they would be required to reduce costs by 66% or grow revenues by 44% — a goal beyond the scope of most banks in the current climate.
A problem of low Return on Equity (peripherals around 2%, average eurozone banks below 6%) and high exposure to government loans is not solved in two years, but deposits have stabilized and banks have sold large packages of toxic real estate assets. That does not make the sector “totally healthy”.
The stress tests of 2014 will be very demanding and assume , among other risks:
- An adverse scenario of falling GDP in Europe of 0.7% in 2014, -1.5% in 2015, plus a 21% drop in house prices, added to increased inflation.
- Losses in the portfolio of sovereign bonds from increases in their risk premiums. Increases of 150 basis points in European premium or 200bps in the US sovereign bonds. Assumes a drop of 6.5% off in Spain, for example, 6% in France and 7.6% in Italy and 4.4% in Germany in the 5-year bonds.
- Possible 25% depreciation of the Hungarian and Polish currencies , and 15% of the Czech, Romanian or Croatian.
Although these may seem aggressive estimates, the expected impact on banks is relatively small .
However, do not forget that these exercises are theoretical and, like everything else, reality often shows unexpected effects. But the exercise is important.
Do not expect the credit will grow dramatically because banks pass the theoretical examination of the stress tests.
Although the level of private credit has begun to recover slowly, with an expected growth of 0.5% to the private-sector, 4.4 billion euros in 2014 -, the European Union remains, by far, the most bloated financial system in the OECD.
- European banks are the most intervened, regulated ans State-controlled of the OECD. Not only due to the weight of public sector banks, but because of the disastrous intervention in the M&A and divestment processes , with governments pushing to lend at all costs. let’s not forget the “crowding out” effect of government debt versus households and businesses, encouraging the purchase of sovereign debt through regulation, as explained here and in English here.
- In Europe the banks finance 80% of the real economy , while in the U.S. is about 30%.
- Total assets of the banking system of the euro area accounted for 349% of GDP in 2013 . A reduction of 12% since 2008. Much higher than the U.S. or Japan (Chart courtesy of Merrill Lynch) figure. While it is true that part of it is because European banks have more deposits, it shows a bloated banking system.
- Additional credit expansion is not the solution , as we forget where we came from … A brutal credit growth since 2001, as the graph below shows (mdbriefing.com). Morgan Stanley estimates that European banks have sold or refinanced only between 20 and 25% of the 700 billion of non-performing loans that the regulation required them to urgently address in 2014.
Banks cannot drink and sing at the same time. It is impossible to strengthen balance sheets, avoid taking excessive risks while lending like 2008 just because we think that credit is the solution. First because they can’t and second because they shouldn’t.
The stress tests of 2014 are not the same as those of 2011. European banks have improved. Non-performing loans are expected to be reduced in 2014 while operating profit is estimated to be up 4.2% after three years of decline . The risks still exist, but it is not as severe as it was in 2011. Forcing to lend at any cost is a great danger.
Liquidity injections by the ECB do not solve a key problem. Where do we put all that money? Europe has an average of 25% industrial overcapacity. When I asked on CNBC a senior manager of the ECB where they thought they were going to invest 400 billion euros of TLTRO, he failed to give me a single key sector where those funds would be deployed.
Credit is growing again, but it should not reach the levels of 2004 to 2010 again. As Von Mises said, “no one should expect that any logical argument or any experience could ever shake the almost religious fervor of those who believe in salvation through spending and credit expansion.”
I’m afraid that with negative deposit rates, liquidity injections and stimuli, we aim to re-ignite the credit bubble before the European banking system recovers its strength. Then, when it bursts, we will surely blame the ‘free market’… and de-regulation.
Important Disclaimer: All of Daniel Lacalle’s views expressed in his books and this blog are strictly personal and should not be taken as buy or sell recommendations.