The Yield Curve of Bonds as an Indicator of Recession or GDP Growth – Analysis Based on Data from the USA and the Czech Republic in 2000–2024

This bachelor thesis examines the predictive power of the yield spread between long-term and short-term bonds on real GDP growth and upcoming recessions in the USA and the Czech Republic during the period 2000–2024. The predictability of real GDP is analyzed using the OLS method. For the relationship with upcoming recessions, the probit model is employed, where the dependent variable is the NBER/OECD-based recession indicator. Both sections utilize data on the yield spread between 10-year and 3-month government bonds.The results show that the yield spread had significant predictive strength for GDP growth in both countries from 2000 to 2020, with a lead time of several quarters. However, this relationship underwent substantial changes throughout the period. For 2020–2024, the relationship could not be estimated due to insufficient data. Due to the non-stationarity of the yield spread, the analysis identified a relationship between changes in the yield spread and future recessions rather than a direct link. The findings suggest that changes in the yield spread have a lagging (following) nature as a recession indicator, rendering them unsuitable for recession forecasting. Despite this limitation, the study extends the existing literature with more up-to-date data for the USA and provides a rare analysis of both relationships for the Czech Republic. The conclusions also highlight the need for further analysis to explore these relationships without the constraints imposed by data characteristics.