Will EU Fail Stress Tests?: Kashyap, Schoenholtz, Shin

(Corrects probable timing of EBA’s Europe-wide stress tests in second, fourth and final paragraphs.)

March 15 (Bloomberg) -- The Federal Reserve’s 2012 stress tests of U.S. banks suffered from some of the same weaknesses as the ones it conducted last year.

We hope the same won’t hold true of the European Banking Authority, which has delayed another round of exams of the region’s banks until 2013, after failing to credibly carry out its supervisory mandate in 2011.

The EBA’s swing-and-a-miss followed similar disappointing performances in 2009 and 2010 by its predecessor in this role, the Committee of European Banking Supervisors. Now, euro-area banks are deleveraging, deepening the recession by starving European companies of financing. They also lack the capital to support an eventual recovery in the region. Until the capital shortfall is measured and addressed, there will be no end to the twin sovereign and banking crises in the euro area.

When the EBA gets a fourth turn at bat next year, the authority can -- and should -- make use of a growing body of knowledge and experience regarding stress tests. And supportive actions by the European Central Bank and the fiscal authorities may raise the chances of success.

The stakes are high: Another stress-test failure would undermine the fragile consensus on policy efforts to contain the crisis, feeding a run on the region’s weaker banks.

What would a successful stress test require? Our research, conducted jointly with David Greenlaw, highlights two prerequisites for success. First, it would need the effective application of macroprudential principles, which means looking at the entire financial system.

Capital Needs

A stress test should aim to prevent fire sales, credit crunches and systemic defaults. It should assess capital needs based on the state of the system as a whole, taking account of the effect of spillovers at one bank on the behavior and well-being of others.

Most importantly, in contrast to the usual exclusive focus on credit risk and the riskiness of bank assets, an effective stress test should attend also to the system’s liabilities -- such as wholesale funding -- that may be vulnerable to a run. Guidance resulting from the tests should include recommendations on the capital levels needed to prevent (or end) deleveraging, which amplifies adverse economic shocks. In this respect, even the Fed’s stress-test results, released March 13, fall short by taking a-one-bank-at-a-time approach to capital needs and paying too little attention to funding needs.

The second prerequisite is a credible strategy to make up the measured capital shortfall. Ideally, the additional equity should come from the private sector. When they are credible, stress tests can reopen private sources of capital to banks. The classic example is the 2009 test of U.S. banks by the Federal Reserve. Yet, no stress test is likely to be credible without an ample public backstop in the event that new private capital isn’t forthcoming. Lacking such a promise of public funds, policy makers cannot reveal that a bank is insolvent, because such an announcement would trigger an instant run.

For the past three years, Europe’s stress tests have failed both of these requirements. Lacking a credible public backstop, the EBA was naturally reluctant to reveal any large capital shortfall. When it first published the 2011 test results in July, the EBA measured the gap at a mere 2.5 billion euros ($3.3 billion). Dexia SA, the French-Belgian lender that failed after a run less than three months later, received a clean bill of health.

Accounting Tricks

Later in 2011, the EBA, using the same stress-test data, raised its estimate of the system’s shortfall to 115 billion euros. Yet, amid the euro area’s poor economic outlook, this revision failed to convince private suppliers of bank capital. As a result, only a few banks moved to bolster their balance sheets. Instead, in a fundamental violation of macroprudential principles, the EBA allowed banks to address their capital shortages in part through accounting tricks that lowered risk-weights on assets and through deleveraging by shedding assets.

The greatest failure may have been the EBA’s lack of attention to bank-funding issues. Compared with banks in other industrialized countries, European institutions are particularly dependent on wholesale funding, which can be impossible to roll over in a crisis. Within the narrower euro area, most of this wholesale funding is cross-border in nature. The result is an explosive mix in the peripheral countries, where many banks hold portfolios of domestic assets while relying on short-term funds from abroad.

It should come as no surprise, then, that a wholesale “bank trot” is well under way in the region. For many banks in the euro-area periphery, the interbank fund market has become inaccessible, forcing reliance on the ECB as the lender of last resort. The enormous market impact of the central bank’s three-year unlimited funding operations in December and February highlights how liquidity-starved these institutions had become.

Instead, much of their funding has flowed to banks in the core euro-area countries, whose central banks recycle them through the ECB. (In doing so, countries such as Germany are accumulating enormous -- and politically controversial -- claims versus the euro area’s central banks. As of January 2012, the Bundesbank’s claims approached 500 billion euros, or almost one-fifth of Germany’s annual gross domestic product.)

To be sure, the science of stress testing is in its infancy, and there is major scope for improvement, even where tests have been more effective than in Europe. For example, following its 2011 stress test, the Fed allowed a distribution of capital by some banks at a time when the banking system as a whole still suffered from a capital shortfall. And this year’s U.S. tests still lack an intimate connection between the “living wills” that banks must promulgate and the estimation of their capital needs. The lack of attention to liquidity concerns also persists.

Stability Mechanism

The EBA, however, can’t afford a fourth strike. Euro-area policy makers are soon expected to discuss enlarging the buffer or firewall when the European Stability Mechanism goes into operation in July. If they reach consensus, the availability of a public backstop will allow the EBA to assess euro-area banks aggressively and insist on capital increases to offset any shortfall.

If European authorities apply basic macroprudential lessons, their next stress test is likely to be far more effective, allowing supervisors to help end the crisis. Otherwise, the run on weaker euro-area banks will probably continue.

(Anil K Kashyap is a professor of economics at the University of Chicago Booth School of Business and a contributor to Business Class. Kim Schoenholtz is a professor of economics at New York University’s Stern School of Business. Hyun Song Shin is a professor of economics at Princeton University. The opinions expressed are their own.)

Read more opinion online from Bloomberg View.

To contact the writers of this article: Anil K Kashyap at; Kermit Schoenholtz at; Hyun Song Shin at

To contact the editor responsible for this article: Max Berley at

    Before it's here, it's on the Bloomberg Terminal.