A model for forecasting and policy analysis in Pakistan: The role of government and external sectors
- The State Bank of Pakistan uses a forecasting and policy analysis system as a tool for monetary policy analysis.
- This project extended the current model to add analysis of fiscal policy and the external sector.
- Adding these areas for analysis to the model is important because fiscal policy exerts an important influence on the formulation of monetary policy in Pakistan, and analysis of the external sector needs to be developed further to account for the lack of integration of Pakistan’s financial markets with global markets.
The State Bank of Pakistan uses a forecasting and policy analysis system as a tool for monetary policy analysis. This is a basic dynamic stochastic general equilibrium (DSGE) model of a small open economy. Our project developed the next generation of the model to meet the central bank’s needs.
A key object of the project was to extend the model to add fiscal and external sector blocks. The new fiscal block modelled the behaviour of government expenditure, tax revenues, and government debt, and allows for government borrowing from the State Bank of Pakistan, which affects money growth.
We revised the external sector to introduce transaction costs in international borrowing and lending, which weaken the link between the return on domestic assets and the exchange rate adjusted return on foreign assets. To examine the behaviour of the major components of the consumer price index, the general model also distinguished three sectors: core products, food, and oil.
We compared forecasts from the current forecasting and policy analysis system model with those of the new basic model we constructed under this project. We found that the new model performed better at all forecast horizons.
We also compared our model with other commonly used forecasting tools. We made forecast comparisons for real GDP, consumer price index inflation, and the nominal interest rate. Our model performed better than the Bayesian vector autoregressions (VARs) model, which is an empirical model, for each variable at all forecast horizons.
We also compared our model with a hybrid theoretical-empirical model. This attempts to improve forecast performance by combining the forecasts of our DSGE model and the VAR model. The hybrid performed marginally better in forecasting interest rates and inflation but slightly worse in predicting GDP growth. It therefore did not contribute much to improving the overall forecasting ability of our model.