Shortly after the publication of the  2016 EBA stress test results, Viral Acharya, Diane Pierret, and Sascha Steffen (APS) presented a paper that identifies capital shortfalls that are significantly larger than EBA’s shortfall calculations reveal.

Reuters quickly picked up the highlight of the paper: “Using the Fed’s approach, the 51 European banks showed a total capital shortfall of 123 billion euros”

123 billion euros of course is a large shortfall, especially given that the shortfall is (only) 5.6 billion euros if one uses EBA’s own methods.

Below I will attempt to reconcile these two different shortfall amounts: EBA’s 5.5 bn euro shortfall versus the one highlighted in the paper and the press: 123 bn euro.

Cherry picking? Inspired by the U.S. Comprehensive Capital Analysis and Review (CCAR 2016) methodology, APS pick the highest shortfall from four metrics: i) the Common Equity Tier 1 Capital ratio, ii) the Tier 1 Capital ratio, iii) the Total Capital ratio, iv) the Tier 1 Leverage ratio. For each of these four capital ratios, the authors take the maximum capital shortfall over the three years (2016-2018) and derive the final capital shortfall measure taking the maximum of these shortfalls.

This looks like cherry-picking results. Moreover, it makes it hard for the reader to compare shortfalls, given that they originate from different calculations. However, only if one looks at Appendix III, the reader discovers that the shortfalls are driven by the Leverage ratio, which APS set at 4%.

A leverage ratio requirement of 4%. The choice to rely on the leverage ratio is an odd one, because European banks are not yet supervised on a leverage ratio basis. Then again, European banks expect the leverage ratio requirement will be set at 3%, so benchmarking banks against a significantly higher 4% leverage ratio requirement looks like moving the goalpost mid-game.

More interestingly, last week the EBA presented a thorough report on the leverage ratio, which makes a well-reasoned case for sticking at a 3% leverage ratio requirement. Moving to 4% (5%) would lead to a shortfall of 82.9 (268.3) bn euros, see the EBA report:

In addition, moving to a 4% leverage ratio requirement would likely affect lending in the short term, which is not really what Europe needs today.

Like-for-like. The authors appear to assume institutions are the same across countries. The paper is very much oriented towards the U.S., a country with fairly homogeneous institutions. Europe is different. For example, a shortfall of a U.K. or Dutch bank (ignoring Brexit for now, which is not unreasonable given that the measurement date is end 2015) is probably different than a shortfall of an Italian bank. The economies are very different, people differ in their discipline and ability to repay mortgages; asset quality and effective maturities may differ across countries too.

More like-for-like. The authors add all shortfalls together as if each unit of shortfall weighs equally, which it probably does not. To illustrate this point, I added the shortfalls identified by the authors to CET1, and then recalculated the CET1 ratios for the 20 banks that top Table 1 of the paper.

As you can see, the effects of the shortfalls vary across banks.  For example, eliminating the shortfall of N.V. Bank Nederlandse Gemeenten (a bank for Dutch local governments that will unlikely cause any trouble for financial stability) will fire up its CET1 ratio to over 45%. For this bank, the authors identify an important shortfall – one that contributes to the 123 bn euro shortfall figure. However, it is unlikely that eliminating that shortfall will contribute to restoring trust in the EU banking sector.

Abn Amro has a low leverage ratio. If my mind serves me right this is because of derivative trading. Covering the gap of 4.9 bn euro will increase the CET1 ratio to 20.1%, which – with a current CET1 ratio of 15.5% (dipping to 9.5% under fully loaded 2018 adverse scenario) – will unlikely contribute to a more stable EU banking system. ING, Rabo (Coöperatieve Centrale Raiffeisen-Boerenleenbank), RBS, Commerzbank, Groupe BPCE, Banco Santander, Landesbank Baden-Württemberg are all identified by authors as banks with serious shortfalls, but they have current CET1 ratio’s over 12.5% and shortfalls that are less than 3.5% of RWAs. Again, it is not clear to me if closing all these shortfalls will contribute significantly to financial stability.

To conclude. APS report important capital shortfalls, but I think that shortfalls at some banks matter more than at other banks. I say this because banks are currently already supervised on the basis of a risk-weighted ratio, and risk-weighted ratios of European banks have increased significantly since the onset of the GFC. Pushing for a higher leverage ratio requirement, which APS appear to do, may not help. Many banks are already well-capitalized if one relies on risk-weighted ratios.

Pushing for a higher leverage ratio may backfire in the short term: European banks are not very profitable now. This limits their ability to augment ratios. For the long term, I support higher leverage ratios. In the short term, however, European banks are … stuck; stuck in a pool of poor performing assets, stuck in economies that are still not performing well.