Forecasting recovery from COVID-19 using financial data: An application to Vietnam

Area of Science:

• Economics
• Econometrics
• Financial Markets

Background:

• The COVID-19 crisis significantly impacted global economies, necessitating accurate Gross Domestic Product (GDP) forecasting.
• Small, export-oriented countries face unique challenges in economic recovery and require specialized forecasting tools.
• Conventional forecasting methods often lag in capturing real-time economic shifts during crises.

Purpose of the Study:

• To develop and validate a new methodology for nowcasting and forecasting GDP evolution in small, export-oriented economies during the COVID-19 crisis.
• To assess the utility of industry-level financial indexes in predicting economic recovery trajectories.
• To compare the proposed method's performance against established international forecasts.

Main Methods:

• Exploiting variations in industry-level financial indexes during the early stages of the COVID-19 crisis.
• Relating financial index fluctuations to expected crisis duration per industry, assuming COVID-19 as the primary shock source.
• Integrating industry-level GDP trend deviations with real-time financial data for predictive modeling.

Main Results:

• The developed methodology produced GDP predictions for Vietnam that were more optimistic than those from the International Monetary Fund and other international forecasters.
• The forecasts generated by the new method more closely aligned with the realized GDP figures.
• The study highlights that early signs of better-than-expected economic performance were present in stock market data but overlooked by traditional forecasting approaches.

Conclusions:

• The novel methodology effectively utilizes financial market data for more accurate and timely GDP nowcasting and forecasting during crises.
• Financial indexes offer valuable leading indicators for economic recovery, particularly in open economies susceptible to global shocks.
• This approach provides a superior alternative to conventional methods for predicting economic trajectories in the face of unprecedented events like the COVID-19 pandemic.