- September 20, 2013
- Posted by: admin
- Category: mefmi-fellows-development
Primary and secondary bond market data and Ordinary Least Squares (OLS) and Generalized Auto Regressive Conditional Heteroskedasticity (GARCH) techniques were used to investigate the effect of the benchmark bond programme on the development of the yield curve in Kenya from the year 2000 to 2016. Both simple and multivariate OLS showed that demand for bonds as measured by subscription rate and bid to cover ratios at the primary market were significant determinants of the level and direction of two and five-year benchmark bond yields, and not 10-year bonds.
The GARCH results indicated that volatility persistency was highest in the 2-year benchmark bond yield, followed by the 5-year and then 10-year yield. In addition, the coefficient for volatility clustering was significant at 1% level, only for the 2-year yield. These outcomes indicated evidence for the 2-year benchmark yield being weak form efficient and the possibility that the medium-long term bond market was on course towards semi-strong efficiency. The findings provide evidence of positive impact of the benchmark bond programme on the development of the yield curve in Kenya thus providing insights for the need to continue strengthening and enhancing the programme to include larger bond issuances and liability management operations. Initiatives to increase demand of bonds at the primary market and level of trading in the secondary market which affect the direction and volatility of yields of bonds should be encouraged and enhanced. The results also provide lessons for the larger MEFMI region to continue supporting reforms for increasing the efficiency of the bond market.
Benchmark Bond Programme and Yield CurveDevelopment in Kenya_Leonard Thotho