Addressing the limitations of forecasting banknote demand

Central banks need to forecast banknote demand. It determines the number of notes they need printed and the future distribution network required. Yet forecasting demand is an inherently complex problem - banknotes are anonymous bearer instruments and so many of the sources of demand are difficult to research. This paper sets out a framework for identifying and assessing drivers likely to influence banknote demand. It presents, for the first time, the findings from an econometric model, looking at the past relationship between demand for Bank of England notes and a range of economic variables and cash industry statistics, to help forecast future demand. But this approach has its limitations. There will be determinants of demand not included in the model. Furthermore, what is to say that past relationships will hold into the future? Perhaps we are now approaching a point of inflection - a paradigm shift in the demand for cash that causes the pre-existing relationships to break down. To account for this, central banks must continue to research cash demand, its current and future drivers, and how significant they might be going forward. They must look for leading indicators that suggest a break with the past, and attempt to understand how, and when, the impact of technological change may significantly change the trajectory of cash use. This paper will set out a structure for capturing all of this information and using it to make judgements on the future of cash. Whilst it might improve central bank’s forecasting capability, and thus the basis for policy decisions, it will not eliminate all uncertainty. Therefore, central banks must retain flexibility, and ensure the wider cash industry does as well. There is a future for cash but we must constantly be alert to events that might change what that future looks like.