This paper investigates the link between jumps in the exchange rate process and rumors of central bank interventions. Using the case of Japan, we analyze specifically whether jumps trigger false reports of intervention (i.e. an intervention is reported when it did not occur). Intraday jumps are extracted using a non-parametric technique recently proposed by Lee and Mykland (2007) and by Andersen et al. (2007), and later modified in Boudt et al.(2008). Rumors are identified by using a unique database of Reuters and Dow Jones newswires. Our results suggest that a significant number of jumps on the YEN/USD have been falsely interpreted by the market as being the result of a central bank intervention. The paper has policy implications in terms of central bank interventions. We show that in times where the central bank is known to intervene, some investors may attach a lot of weight to central bank interventions as a source of exchange rate movement, leading to a false "intervention explanation" for observed jumps.
|Publication status||Published - 2012|