Exchange rate volatility and central bank interventions

This paper studies the impact of Swiss National Bank interventions, and news about these interventions, on the intraday volatility of the Swiss franc - U.S. dollar exchange rate. It extends the existing literature by characterising the the impact of different aspects of central bank interventions, like direction, size, frequency and time of intervention, on exchange rate volatility. Briefly, the paper finds that the effect of intervention on volatility varies depending on how volatility is defined. Interventions decrease volatility contemporaneously but this effect is reversed in the two hours afterwards. This relationship is symmetric with respect to the direction of the intervention, whether they be buy and sell interventions or with-the-wind and against-the-wind interventions. Analysis of the volatility and intervention size relationship finds that as we move from small to large interventions, the larger interventions tend to increase volatility relative to small interventions. The frequency of interventions has a small but positive impact on volatility, and this is underscored when the analysis is done by splitting the sample into low, average and high frequency interventions. The interaction between intervention size and intervention frequency results in a small positive effect on volatility for the squared return measure and the absolute return measure and a negative effect for both the realised volatility measures this effect is negative. As before the effect of the timing of the intervention varies with the volatility measure. The relationship is different for interventions at different times of the day. For the two realised volatility measures 9am interventions reduce volatility while for the other two measures the significant coefficients have an overall positive effect increasing volatility. 2pm interventions decrease volatility for both the squared return measures but increase volatility for both the absolute return measures. Reuters reports of sell interventions have a significant and lagged negative effect on volatility for the squared return measure and both the absolute return measures.