Why are institutions not backtesting their allowance calculations?

Banks and credit unions backtest their allowance for loan and lease losses (ALLL) in order to compare actual results for a defined period to the results forecasted by a model for the same period. This allows institutions to monitor model performance and adjust or revise it over time. It also helps institutions defend their ALLL methodology to the board, auditors and examiners. 

But according to a poll on a recent webinar, Backtesting Your ALLL Methodology, almost 50 percent of banks and credit unions aren’t currently backtesting their ALLL. The same pool of over 200 bankers voted on reasons why they weren’t backtesting, finding that 36 percent are unsure of how to do it, while another 37 percent find the time it takes to be a constraint. Over 25 percent point to data issues. 

When attempting to backtest the ALLL, banks and credit unions can face data challenges in two key areas. First, many may not have sufficient data stored to properly backtest. But even institutions that do have the proper data stored may not be able to easily access or work with that data. Not only could this lead to time constraints, another primary challenge referenced above, it could prevent the ability to backtest effectively. 

Implementing an automated system, like Sageworks ALLL, that can archive data and make it accessible can help save time and enable bankers to effectively backtest their ALLL methodology.

For specific examples and report samples on how to backtest, download the whitepaper, Backtesting: Measuring the Effectiveness of ALLL Methodologies.