The thesis empirically examines and assesses the term structure dynamics in the context of the European Central Bank. The work is based on previous research in the setting of the US Federal Reserve Bank. Within a multifactor term structure model, particular focus is put on the interplay of the ECB, its monetary policy and the influence on the yield curve. More specifically, the thesis develops a continuous time jump diffusion model that explicitly accounts for the target rate changes of the ECB as an observable fourth factor.
In order to achieve the goal of analysing and modelling the impact of the ECB on the term structure of interest rates and vice versa, the thesis divides the theme into six different chapters. The first chapter introduces the relevance of the analysis and the importance of the topic. Throughout the second chapter, significant mathematical concepts as well as an overview on affine term structure models with jumps and their estimation has been presented. As the target rate of the ECB has been included as an explicit fourth factor in the analysis, a basic understanding of the ECB and the conduct of monetary policy is vital in the understanding of the model setup. This has been the topic of the fourth chapter. Next to the general model setup also the relevant dataset as well as the estimation method have been presented in chapter five. The following part is then attributed to the estimation outcomes as well as the resulting implications for the modelling of the European term structure dynamics in conjunction with the policy of the ECB. The sixth chapter concludes.
Summing up the outcomes of the different empirical analyses conducted, the following aspects can be highlighted:
• The direct inclusion of the target rate as a fourth factor into the model seems to significantly improve the cross-sectional pricing performance in comparison to term structure models with only latent factors. This is true for the short end of the yield curve, where the inclusion of the target rate appears to cover the enhanced volatility as well as the seasonality around central bank meeting days. Yet, the long-end fit is also improved.
• The consideration of an aggregate macroeconomic factor is also valuable with regard to the fit of the model and the pricing performance. This is in particular true for the longer end due to the relatively long-term focus of this variable. It thus also helps to assess the overall stance of the economy as well as market expectations with respect to future economic developments.
• Policy inertia and interest rate smoothing play an important role in the policy of the ECB. Thus, it should also be captured within the modelling of the term structure dynamics in the Euro area.
• The general form of the model, i.e. the inclusion of the target rate in form of a jump process, together with a relatively flexible specification of the volatility and the inclusion of a macroeconomic factor is heading into the right direction with respect to future research on European term structure dynamics. This finding has been underlined by the specific volatility structure found as well as by the increased level of volatility around central bank meetings in the data.
• All aspects are finally revealed in the estimated policy rule that clearly outperforms different versions of the Taylor rule and thus noticeably improves existing research in this area.
On the whole, the conducted analyses have resulted in valuable insights on the term structure dynamics in the Euro area, the interplay with the monetary policy of the ECB and the modelling of the yield curve in this environment. In order to enhance research in this area, several directions to advance and continue from here are worth highlighting.
Miriam Begtasevic
ECB EZB European Term Structure Interest Rate Jump Processes Term Structure