Europe’s latest bank stress test was flawed, and dozens of the region’s lenders, including Deutsche Bank AG and BNP Paribas SA, aren’t sufficiently capitalized to improve the economy’s anemic growth or withstand a repeat of the 2008 financial crisis.
Those are the conclusions of analysts at the Danish Institute for International Studies and data from Keefe, Bruyette & Woods Inc. showing what would have happened if the European Central Bank had applied a leverage minimum that will be introduced next year. A study by the Brussels-based Centre for European Policy Studies showed Deutsche Bank and BNP Paribas above the cutoff, while 28 other banks that passed the stress test failed.
The new standard requires banks around the world to have capital equal to 3 percent of total assets, complementing a system that weights them for risk. If the ECB had used that yardstick and demanded the highest quality capital, 12 big European banks that passed the stress test would need to raise an additional 66 billion euros ($82 billion), according to Jakob Vestergaard, a senior researcher at the Danish institute.
“Relying on risk-based measures only isn’t enough because it’s always what we thought wasn’t risky that ends up blowing up during a crisis,” said Vestergaard, who examined data collected by the ECB at the request of Bloomberg News and has published papers on leverage. “The ECB wanted to appear tough, but it still couldn’t show big German, French banks as undercapitalized for political reasons.”
The ECB didn’t subject bank leverage ratios to the stress test’s adverse economic scenario because European lenders only have to report those numbers on an informational basis starting next year, a spokeswoman for the central bank in Frankfurt said. The new international standard approved by the Basel Committee on Banking Supervision won’t be fully binding until 2018.
When it released test results on Oct. 26, the ECB provided leverage data that showed 14 lenders, including Deutsche Bank, were below the 3 percent minimum. Three more fell short after the central bank’s asset-quality review determined how many loans should be considered nonperforming. Combining the results of the independent studies, almost three times as many banks would fail the stress test if the leverage standard were used.
The ECB said at the time that it might be misleading to use leverage in a stress test because the central bank assumed balance sheets would remain the same over the next three years. That assumption would affect other test results the same way and isn’t a reason to avoid using a simple leverage ratio based on total assets, Vestergaard said.
The ECB used banks’ capital ratios based on their own models of risk assessment to determine which firms could survive a simulated recession. The models have been criticized by some regulators for underplaying risk and overstating financial strength. U.S. stress tests, which helped restore confidence in the nation’s banks after the financial crisis, have used a simple leverage threshold along with one based on risk.
A 2012 study by the Bank of England, which looked at 100 big lenders around the world and 8,500 U.S. banks of all sizes, showed that the simple leverage ratio was a better predictor of bank failures than the one relying on risk assessments.
When a bank has too much leverage -- financing its business mostly with debt -- it can go bankrupt if the value of its holdings drops even a little. A firm with capital equal to 2 percent of total assets could see its equity wiped out when prices of those assets fall an average of 2 percent. The lower the ratio, the weaker the bank.
“In some ways, the leverage ratio is more important for a stress test because it works better at telling you which banks are weak,” said Morris Goldstein, a senior fellow at the Peterson Institute for International Economics in Washington. “There are some very large European banks with very low leverage ratios, and that’s not good for Europe’s financial stability.”
Banks have said a leverage rule that doesn’t take risk into account is too simplistic.
“If you look at a bank’s total assets, you learn nothing about the quality of those assets,” said Ronald Weichert, a spokesman for Frankfurt-based Deutsche Bank. “You cannot equate cash with subprime mortgages. Not only do leverage ratios fail to tell us what we need to know -- leverage restrictions may also encourage risky behavior.”
In the three studies conducted since the ECB published the results of its stress test, the analysts used different methods and estimates because the banks and the ECB didn’t provide sufficient information to calculate the leverage ratio under the adverse conditions of the stress test. In all, 42 banks failed at least one of the reviews in addition to eight asked to raise more capital by the central bank.
Deutsche Bank, Germany’s largest lender, and BNP Paribas, France’s biggest, had leverage ratios below 3 percent under the ECB’s adverse scenario, according to the studies by Vestergaard and KBW, an investment bank that focuses on financial institutions. Both used banks’ figures for common equity Tier 1 capital, which includes shareholder equity and some elements such as goodwill that won’t be counted in the final version of the new Basel standard.
Societe Generale SA, the second-biggest bank in France, and No. 4 Groupe BPCE also would fall below 3 percent, according to Vestergaard and KBW.
ING Groep NV, the biggest Dutch bank, fell short in the analysis conducted by KBW’s London-based banking team, which also was done at the request of Bloomberg News. Amsterdam-based ING had a leverage ratio of about 3 percent in Vestergaard’s review.
Raymond Vermeulen, a spokesman for ING, said “there continues to be regulatory uncertainty on the calculation methodology for the leverage ratio” in Europe, which could change where final numbers will end up. The Netherlands has gone further than the global standard and is asking the country’s biggest lenders to meet a 4 percent leverage minimum based on total assets by 2018.
Spokesmen for BNP Paribas, Societe Generale and BPCE declined to comment about the findings.
The Centre for European Policy Studies, which has done extensive research on banking issues in Europe, identified 34 lenders that would fall short of the leverage target in a simulated recession, including six that failed the ECB test, according to a study of 130 lenders it published on Nov. 10. The 34 banks would face a capital shortfall of 21 billion euros.
The group used an older definition of capital, one that includes more hybrid securities. Its list of banks failing a stress test based on leverage didn’t include Deutsche Bank, BNP Paribas, Societe Generale or BPCE.
“There’s not enough data publicly available,” said Willem Pieter de Groen, a researcher who conducted the analysis. “Even though the ECB has enhanced transparency on the region’s banks with its exercise, there’s still room for substantial improvement.”
KBW, which examined only the 30 banks it covers, used a narrower definition of capital based on national rules allowing exemptions during the transition period before the Basel leverage ratio goes into effect.
The Danish institute, which analyzed the 32 largest banks in 10 European countries, went even further, defining capital as it will be in 2018, when the Basel rule is fully implemented.
Banks lobbied to dilute the new leverage standard as it was being crafted by the Basel committee. While the initial plan was to include most credit default swaps in the calculation of total assets, the industry won concessions allowing banks to net some derivatives exposures. That helps them continue hiding some of their risks, according to Gordon Kerr of advisory firm Cobden Partners in London.
“Even the leverage ratio is being gamed,” said Kerr, who advises governments and investors on crisis management. “They can offset their derivatives and make the total exposure smaller than it really is. We’re approaching another crisis, and too many of these banks are still insolvent in reality.”
The ratio of risk-weighted assets to total assets at the largest banks has fallen over the past two decades, according to a 2012 study by the Organization for Economic Cooperation and Development. The average risk-weighting dropped to 33 percent of total assets in 2008 from 66 percent in 1991. Last month the Basel committee said it was planning to introduce measures to stop the erosion of risk-weightings.
“Have the banks’ assets really become less risky?” asked Vestergaard. “Clearly not, especially because the lowest point was when the financial crisis erupted and losses surged. They’re just gaming the system better and better.”
The continuing undercapitalization of banks hurts the euro region’s efforts to boost economic growth, according to Alberto Gallo, head of European macro credit research at Royal Bank of Scotland Group Plc. Even if the banks can survive, too many are too weak to restart lending to consumers and companies, he said.
“The balance-sheet problem has been resolved in the U.S. through debt restructuring and home foreclosures,” said Gallo, who’s based in London. “In Europe, bad debts are still on balance sheets, not being restructured. That holds back the region’s growth.”