Quantitative PPNR Modelling

Amnon Levy

The Federal Reserve’s annual Comprehensive Capital Analysis and Review (CCAR) assesses the capital adequacy of large, complex US bank holding companies (BHCs), and the practices used to assess their capital needs. As part of the CCAR process, institutions must demonstrate that they have sufficient capital to withstand a severely adverse operating environment and be able to continue operations. So far, institutions have mostly focused on developing quantitative models to describe potential credit losses, with expert judgement being more heavily relied upon for remaining balance-sheet items. However, with the Federal Reserve increasing the required sophistication and precision of CCAR submissions, there is a growing need to gather data and develop quantitative models for the broad set of balance-sheet and income statement items. Specifically, there has been more focus on developing quantitative models for pre-provision net revenue (PPNR), defined as interest and non-interest income, less interest and non-interest expense. PPNR represents a large number of disparate income statement items, ranging from interest income on loans, interest expenses related to retail deposits