A structural model of capital buffer usability | Jan Hannes Lang, Dominik Menno
Banks cut lending sharply rather than let excess capital ratios slip below regulatory buffers even at tiny costs, a Bundesbank paper says.
Source: Deutsche Bundesbank · August 14, 2026 at 5:55 PM · AI-assisted report
Single-sourceKUALA LUMPUR, 15 AUGUST 2026 —
Listen to this article
DomainFork Audio · read aloud
Banks cut lending sharply rather than let excess capital ratios slip below regulatory buffers even at tiny costs, a Bundesbank paper says.
Market Impact
Deutsche Bundesbank economists Jan Hannes Lang and Dominik Menno modelled capital buffer usability in a structural non-linear banking framework. They found stigma costs as low as 0.5 basis points trigger sharp deleveraging when losses hit. “When credit risk materialises and banks make losses, the same small stigma costs prevent them from letting their excess capital ratio fall below the CBR, instead forcing significant deleveraging of up to 10%,” the authors say.
Regulators added capital buffer requirements under Basel III to bolster resilience without immediately restricting credit supply. Unlike minimum capital rules, banks may operate below the buffer during stress but face supervisory scrutiny, dividend curbs and market stigma if they do. Empirical studies have yielded mixed signals on whether these constraints bite; the Bundesbank paper provides a theoretical answer.
“Our model shows that non-releasable macroprudential buffers may fail their macro stabilisation objective in crisis times,” Lang and Menno conclude.
The model calibrates default costs at 1% of assets and long-run credit risk at 50 basis points. Under these parameters, meeting the buffer adds only 1.9 basis points to funding costs while cutting annual bank failure probabilities by 2 percentage points. In normal times, the paper estimates lending would fall by 1-13 basis points while capital ratios rise 0.5-1.3 percentage points and default probabilities drop 1.25-2 percentage points.
“Introducing a CBR in profitable periods increases resilience with little drag on credit supply,” the authors note.
Losses change the picture. With high credit risk, banks that try to absorb shocks instead shrink balance sheets to avoid violating the buffer requirement. The paper quantifies the deleveraging pressure at up to 10% for the credit-risk ranges studied. “This indicates a structural non-releasable CBR is unlikely to fully achieve its macro stabilisation goal,” the economists write.
They suggest reconsidering the mix of releasable versus non-releasable buffers. “Further analysis is needed to determine the optimal composition,” they add.
Related: Bank Negara Malaysia