Page 98 - Special Topic Session (STS) - Volume 4
P. 98

STS566 Iluminada T. Sicat
                  case of extraordinary circumstances such as those arising from natural disaster
                  or  system-wide  financial  bank  run.  Buffer  stocks  are  maintained  for  all
                  denominations of notes and coins, whereas contingency reserves are only for
                  high  denomination  banknotes.  Recently,  the  BSP  Monetary  Board  has
                  integrated the 2 types of prudential reserves into one, and called it simply as
                  “Buffer stock”, for easier monitoring and to be consistent with the practice of
                  other central banks. Currency demand forecasts also consider the lifespan of
                  the banknotes/coins and the need to replace them whenever the quality of
                  the banknotes and coins are no longer fit for circulation. Notes and coins in
                  circulation must be of a certain quality, in order to promote easier detection
                  of the authenticity of the notes, hence enhance the integrity of the currency.
                  Because the Philippines consists of many islands surrounded by water, climate
                  is humid, and wet markets exist, the Philippine banknotes, particularly, lower
                  denominated ones have a shorter lifespan, hence faster replacement of notes
                  in  circulation.  Over  the  years,  the  BSP  has  improved  the  Philippine  peso
                  substrates particularly for lower denomination banknotes in order to lengthen
                  their lives. The aggregate currency demand in the Philippines is derived by
                  summing the projections coming from these 3 variables.

                      Currency demandt = ƒ (growth, mandated buffer, replacement rate)t

                  To  calculate  how  much  banknotes  and  coins  will  be  ordered,  the  end-of-
                  period inventory or the stock on hand is deducted from the projected annual
                  currency demand.

                                     Currency Order = Demandt – Inventoryt-1

                      Outcome of the currency forecast exercise indicate that the Philippines will
                  need an average of about 4-4.5 billion banknotes annually for the next 5 years.
                  Meanwhile, forecasts on denominational mix are based on historical trend.
                  Error-correction  term  is  incorporated  to  adjust  for  substitution  of
                  denominations due to unavailability of requested denomination. The error-
                  term  also  captures  change  in  denominational  mix  arising  from  financial
                  innovations or alternative payment infrastructure.
                      Regional Currency Distribution. Having forecasted the aggregate demand
                  for the entire country, the next step is to allocate the regional distribution
                                                                                      6
                  according to each 23 currency distribution centers of the BSP. ARIMA  (auto-
                  regressive integrated moving average) model using time series data is used


                  6  Auto-regressive integrated moving average (ARIMA) models are a form of statistical time
                  series which seek to capture the dynamic stochastic aspect of time-series data through a
                  lagged variable structure while abstracting from more predictable features in the data (such as
                  trends and seasonality). Univariate (single vector) ARIMA is a forecasting technique that
                  projects the future values of a series based entirely on its own inertia. Its main application is in
                  the area of short-term forecasting requiring at least 40 historical data points.
                                                                      87 | I S I   W S C   2 0 1 9
   93   94   95   96   97   98   99   100   101   102   103