Cash Management Agility in Banking: Balance Management and Predictive Forecasting

Abstract

Cash management in banking must keep enough cash for service while limiting idle funds, and transaction patterns keep shifting across units and time. This study examines how cash management staff at a national bank in Indonesia, anonymized as Bank X, run predictive forecasting, deviation monitoring, corrective action, and balance management, and how information technology supports these decisions. The study uses a qualitative case design with in-depth interviews of nine informants in cash management functions. The case covers the Malang, Kediri, and Jember area. The researcher checked interview accounts against daily balance records, which are confidential and not reported. The records matched the accounts. Informants described short- and long-horizon forecasts, daily monitoring, balance consolidation across locations, and technology support such as dashboards, machine learning for ATM cash, and robotic process automation. The head office sets balance limits as monthly average balances, and units manage balances against those limits. Informants perceived that technology speeds up their response, and they check exceptions with their own judgment. The units do not use one common measure of deviation, and staff do not always record corrective actions in one standard way. The study proposes organizational agility in cash management as a repeating forecast-monitor-correct-rebalance cycle. Banks can set common deviation limits, record the cause and size of each correction, and review recurring forecast errors without giving up operational flexibility.