|
|
Sabine Lautenschläger says:
”I remember the often passionate discussion in the Basel Committee on Banking Supervision (BCBS) about balancing risk sensitivity and simplicity. Not all of us believed in risk sensitivity as a basis for capital requirements.
”So first of all, I am glad that we still have a risk-sensitive capital framework, although restricted by important backstops.
”I am convinced that risk sensitivity is the only way to give banks effective incentives. Bigger risks need to be accompanied by higher capital buffers. Lower risks need smaller capital buffers. If we were to de-link capital requirements from risks and, therefore, potential losses, we would have to deal with unhealthy consequences.
”Although I am convinced by the idea of making capital requirements more risk-sensitive, I acknowledge that estimating actual risk weights is very hard. Thus, it was very important to combine greater sensitivity with backstops. So, with Basel III we introduced the input and output floors, which will limit the influence of modelling choices. And we prohibited the most complex model choices for some specific asset classes.
”And then there is the leverage ratio, of course, which serves as an additional backstop to the risk-sensitive capital requirements.
”Overall, I think we managed to strike a balance by combining a risk-sensitive approach with solid backstops.
”But did we manage to keep the framework simple? Well, it is not actually simple; that much I admit. But we must remember that Basel III is targeted at a financial sector that is very complex. It is a grave mistake to believe that there are simple solutions to complex problems.
”Besides this general point, there are several issues which should and could be dealt with, or discussed at the BCBS. Work can be done to improve joint understanding of emerging risks in the banking sector, to exchange information on standards and best practices on new or increasingly relevant topics, and to work on a minimum convergence of supervisory practices. And if I am informed correctly, many of the topics I will mention can be found in the 2019 work programme of the BCBS: