Building stress-testing models helps a bank reduce risk and improve capital planning

Regulators and shareholders alike are showing increasing interest in PPNR projections. Robust quantitative models for forecasting balances and revenues can help banks not only improve regulatory compliance, but also generate more efficient, flexible, and reliable base and stress forecasts and rapidly test what-if scenarios. In addition, the models enable banks to strengthen their analytical and capital-allocation capabilities, improve the accuracy of capital requirements, align forecasts and metrics for stress-testing and budgetary processes, and free up management time for value-creating business activities.

The bank needed to develop expertise in PPNR modeling at speed and put in place supporting organizational structures and processes. So it asked us to help it build a first set of models and develop the capabilities it needed to continue building more. It was looking not just for analytical and modeling support, but for a pragmatic approach that would deliver results quickly and a top-management perspective that would ensure the effort generated business value for the whole organization.

McKinsey formed a ten-strong team that included a senior analytics expert and professional modelers as well as consultants, who acted as “translators” helping technical specialists and business leaders to understand one another. We began by helping the bank set up a war room to manage this large-scale advanced-analytics risk project. We also helped it appoint a senior external adviser and hire a handful of other risk professionals to form the core of its new stress-testing group.

The bank’s goal was to develop a separate model for each product in its portfolio—some 100 models in all. To forecast balances in mortgages, for instance, it needed different models for different loan durations. Our team was engaged to help build the first 15 to 20 models from the full suite. To be as accurate as possible, the models included variables linked to macroeconomic factors—including interest rates, GDP, unemployment, and the potential impact of Brexit—and not just industry market shares.

The first phase of the project involved cleaning up data, and here the McKinsey team stumbled on an unforeseen difficulty. Because much of the bank’s growth had come through acquisitions, product portfolios were split across multiple legacy IT systems at multiple locations; for instance, there were 15 real-estate databases. The need to assemble and integrate data from diverse sources meant that a task scheduled to take a week ended up taking a month.

With phase two of the work starting late, the deadlines looked to be in jeopardy. However, the project team was able to shorten the model build by using a proprietary tool to run multiple regressions in parallel rather than sequentially, so that the work took just four weeks to complete, rather than the six to seven expected. The team found, for instance, that they could get to a model specification in a couple of hours, not the five days originally envisaged.

By engaging representatives from almost every function in the development process, the team ensured that the resulting models would be robust, include all relevant variables, and be rapidly adopted across the organization.

By the end of the three-month project, the first set of models were in place and covered 20 to 30 percent of the bank’s balance sheet. Similar projects at other banks had taken 18 months to cover 80 percent of the balance sheet. Now that the client has the expertise and resources to build models for itself, it is working on a further phase that will take coverage to 60 to 70 percent of its balance sheet.

Using the PPNR models has given the bank a clearer view of interactions between different parts of the balance sheet and enabled it to make smarter decisions. In a stressed environment, it might previously have taken on extra debt to secure sufficient funding, for instance, but deeper insight into its loan-to-deposit ratio showed that it could attract more deposits than conservative estimates had suggested. By avoiding taking on unnecessary debt, the bank made substantial savings—one among many sources of value creation from better decision making.

Overall, the models have generated valuable insights that will inform the bank’s future budgeting, strategic planning, and business actions. One important example is the bank’s use of the models in scenario testing to forecast the impact of Brexit on mortgage and deposit balances.

The bank regards the models as an enormous success, and several of its project leads have since been promoted. One executive commented drily that budget estimates that had taken six to eight weeks now took only five days—and that was one day to run a model, four days to do the presentation.