OpinionCapital requirements

Why capital buffers in Europe vary so widely and what to do about it

Additional capital requirements for the German credit industry should be clearly justified. Unfortunately, BaFin is reluctant to do so.

Why capital buffers in Europe vary so widely and what to do about it

How much systemic risk buffer is allowed? 3% for non-financial companies from the home country, as the financial supervisory authority in France prescribes for banks? Up to 1% on all loans, as is customary in Austria? Or even 9% based on the exposure that is backed by real estate from the home country and can be attributed to private individuals, as the regulator in Belgium believes?

Supervisors' assessments of how much additional capital the banking industry needs to hold in relation to risk-weighted assets - and for what - vary widely. Anyone who wants to can recognize actual differences in the European banking system, because real estate markets, lending standards, fixed interest rates, bank resilience and the economic situation differ from country to country. However, when viewed with less benevolence, there is a margin of discretion that national supervisors use differently. If you are skeptical, you may also recognize some arbitrariness in the heterogeneous requirements.

With a systemic risk buffer for residential real estate loans of 2%, Germany is roughly in the middle of the range of what is common in Europe. There may be solid reasons for this. Unfortunately, BaFin's analysis cannot be substantiated. It talks about an „indicator-based analysis“ and a „residential real estate stress test“. The „loss potential“ has „not substantially decreased“ within two years. Aha, then. But how exactly does the financial supervisory authority arrive at this result? What exactly does the model calculation in the stress test look like? That remains a mystery.

Precise justification is also in BaFin's interest

That is a problem. A sensible weighing up of costs and risks is essential for all capital requirements. The risk of a banking crisis should be as low as possible, but the capital costs should not be too high either. Does a supervisory authority that does not have to bear capital costs itself make this assessment appropriately? BaFin itself should have an interest in this evidence.

Of course, every model calculation can be set up differently and every analysis can be conducted differently. Dealing with risks is not an exact science, but leaves room for consideration and interpretation. The more concrete the justification, the more precisely counter-arguments can be formulated. However, this should not be an argument against a comprehensive justification. Rather, a precise analysis facilitates the debate on how much capital banks need.