Abstract

We examine the ability of short rates and yield spreads to forecast the growth in Australian industrial output. We find that since 1990, the short rate has a significant increase in its predictive power for forecasting output growth in many industries. We document this increase. The yield spread, on the other hand, is useful in predicting the growth of industries with a `longer' production cycle, such as manufacturing and wholesale trade. Hence, the predictive power of the yield spread on total GDP, is mainly from its ability to forecast these industries. Our out-of-sample forecasts show that yield spread is a good forecasting device for many industries, particular for output growth over longer horizons.

Highlights

  • The predictive power of yield spreads of interest rates on GDP growth has been well studied in recent years

  • We examined the predictive power of the short rate, the long yield and the yield spread for industry level output growths in Australia

  • We find that firstly, the short rate and the yield spread are useful in predicting output growth for many, but not all industries

Read more

Summary

Introduction

The predictive power of yield spreads of interest rates on GDP growth has been well studied in recent years. By decomposing the total GDP into different industries, we are able to determine which industries in the economy are more predictable by yield spreads, and help us to understand about the forecasting ability of yield spreads in more detail This kind of study is not able to be done across different countries. We wish to examine the forecast abilities for short and long interest rates as well as the difference of the two, the yield spread on industrial level output growth. As a result of this, we find that the predictive power in yield spread on the total GDP is from its ability to predict the output growth in manufacturing and wholesale trade.

Data for industry outputs
Some Reviews
The model and estimation results using the whole sample
Possible regime switches and estimation results in the new regime
Does yield spread provide extra information for prediction
The out of sample forecasting
Findings
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call