Abstract/Description

In this study, Bayesian maximum likelihood estimation of quarterly projections model for Pakistan as presented, as documented in Ahmad & Pasha (2015). Estimation results based on quarterly data from 2001 to 2023 show substantial differences in values of estimated versus calibrated parameters related to aggregate demand, aggregate supply, monetary policy rule and exogenous shock processes. The aim of this study is to compare forecasting performance for key macro variables. It shows that the estimated model provides more precise forecasts in case of headline inflation, real GDP growth, interest rate and exchange rate over 8-quarters forecast horizon. An estimated model for gap analysis and scenario analysis was used. Gap analysis, based on March 2023 data, shows that Pakistan is passing through a recession with overshot exchange rate. In scenario analysis, implications of political instability, climate risks, commodity prices and global financial conditions for next three years’ forecasts of domestic variables under baseline and alternate scenarios were incorporated. The scenario analysis shows that simultaneous realization of assumed risk factors may lead to substantial deterioration of macroeconomic outlook and under current circumstances, using an expansionary monetary policy may lead to substantial rise in inflation and macroeconomic volatility without offering sustainable gains in GDP growth.

Keywords

Bayesian Analysis, Applied General Equilibrium Models, Forecasting and Simulation

JEL Codes

C11; D58; E37

Location

S1 room, Adamjee building

Session Theme

Evolving Dynamics in Inflation, Monetary and Fiscal Policy

Session Type

Parallel Technical Session

Session Chair

Kalim Hyder, State Bank of Pakistan

Session Discussant

Ilfan Oh, Institute of Business Administration ; Karim Khan, Pakistan Institute of Development Economics

Start Date

10-12-2024 3:15 PM

End Date

10-12-2024 5:15 PM

Click the Download button to view presentation slides.

Share

COinS
 
Dec 10th, 3:15 PM Dec 10th, 5:15 PM

Bayesian Estimation of a Pragmatic Model for Monetary Policy Analysis: The Case of Pakistan

S1 room, Adamjee building

In this study, Bayesian maximum likelihood estimation of quarterly projections model for Pakistan as presented, as documented in Ahmad & Pasha (2015). Estimation results based on quarterly data from 2001 to 2023 show substantial differences in values of estimated versus calibrated parameters related to aggregate demand, aggregate supply, monetary policy rule and exogenous shock processes. The aim of this study is to compare forecasting performance for key macro variables. It shows that the estimated model provides more precise forecasts in case of headline inflation, real GDP growth, interest rate and exchange rate over 8-quarters forecast horizon. An estimated model for gap analysis and scenario analysis was used. Gap analysis, based on March 2023 data, shows that Pakistan is passing through a recession with overshot exchange rate. In scenario analysis, implications of political instability, climate risks, commodity prices and global financial conditions for next three years’ forecasts of domestic variables under baseline and alternate scenarios were incorporated. The scenario analysis shows that simultaneous realization of assumed risk factors may lead to substantial deterioration of macroeconomic outlook and under current circumstances, using an expansionary monetary policy may lead to substantial rise in inflation and macroeconomic volatility without offering sustainable gains in GDP growth.