Abstract

This study investigates if Producer Price Index (PPI)-Manufacturing is a good indicator for Consumer Price Index (CPI)/inflation in the Philippines. The dataset used is the time series of both CPI and PPI which year-on-year (YoY) growth rates are computed and observed at a monthly frequency. The time horizon for the series is January 2018 to December 2023. The initial estimation period is from January 2018 to December 2022 and out-of-sample period (forecast evaluation) is 2023. To test robustness, five-year rolling windows, as alternative estimation periods, serve as points of comparison. The statistical time series method employed is Vector Autoregression (VAR) models. Follow-up procedures such as impulse response function (IRF) and forecast error variance decomposition (FEVD) are implemented. We conclude that PPI is a useful indicator for inflation targeting. Statistical modelling has shown the potential of PPI as a leading indicator for inflation. IRF and FEVD results show that PPI precedes CPI instead of the reverse. Dynamic and static forecasts of VAR are more accurate when CPI (non-seasonally adjusted) is specified instead of seasonally adjusted. Static forecasts, in particular, yield Mean Absolute Percentage Error (MAPE) that are very close to the 10% threshold. Dynamic forecasts illustrate that it can follow the actual trend of CPI. This work is very timely considering the impact of high inflation on standards of living on a developing country like the Philippines.