Infectious Disease-Related Uncertainty and the Safe-Haven Characteristic of US Treasury Securities

Using daily data from November 1985 to July 2020, we analyse the impact of a daily newspaper-based index of uncertainty associated with infectious diseases (EMVID) on the level, slope and curvature factors derived from the term structure of interest rates of the US covering maturities of 1 year to 30 years. Results from nonlinearity and structural break tests indicate the misspecification of the linear causality model and point to the suitability of applying a time-varying model that is robust to misspecification due to nonlinearity and regime change. We thus use a dynamic conditional correlation-multivariate generalised autoregressive conditional heteroskedasticity (DCC-MGARCH) framework and the results indicate significant predictability of the three latent factors from the EMVID index at each point of the entire sample, and also provide evidence of instantaneous spillover. Finally, we comprehensively determine the safe-haven characteristic of the US Treasury market by analysing the signs of the underlying time-varying conditional correlation between the level, slope and curvature factors and the EMVID index. Results show that US treasuries with long-term maturities as captured by the level factor are consistently negatively correlated with the EMVID index, i.e., they act as a safe-haven, with the slope factor (medium-term maturities) following this trend since 2007, and the slope factor (short-term maturities) also showing signs of a safe-haven since May of 2020. Overall, the findings provide reasonable evidence to imply that US Treasury securities can hedge the risks associated with the financial market in the wake of the current COVID-19 pandemic.