【興味深い研究内容🔍】Official Japanese Intervention in the JPY/USD Exchange Rate Market: Is It Effective and Through Which Channel Does It Work?🌟BoJ Report No.3📚

The empirical approach of the paper builds on the work of Dominguez and Frankel (1993b) and their (OLS) estimations of daily effects of official Bundesbank and Fed intervention on the DEM/USD exchange rate over the 1982 to 1990 time period.9

Using a standard GARCH time series methodology and a comprehensive list of macro news control variables (measuring the news surprise as the difference between official news announcements and results of surveys of expectations of these announcements conducted by Bloomberg during the days preceding the announcements), the findings of this paper show that for the January 1, 1999 to December 31, 2003 time period, official intervention, whether or not the market is aware of the intervention, exerts a significant same-day influence on the JPY/USD exchange rate.

The analysis does not detect a systematic and significant link between days when there is a rumor of intervention, but no intervention occurs, and the JPY/USD exchange rate.10

For the first quarter of 2004, the analysis shows that neither actual intervention nor rumors of intervention alone have any impact on the JPY/USD exchange rate.

9:The empirical analysis of Dominguez and Frankel (1993b) uses news reports of intervention to classify whether an intervention is secret or reported.
They do not address through which channel effective intervention works.

10:See Dominguez and Panthaki (2007) for an intraday study of perceived intervention.

The results, therefore, suggest that Japanese intervention is effective during the first 5 years of the sample and ineffective during the last 3 months of the sample.

This provides an expost rationale for why Japan intervened as well as for why the interventions stopped.11

Moreover, the results suggest that when intervention is effective, it works through a portfolio-balance channel.
The results do not rule out that effective intervention also works through signaling.

11:See Taylor (2006) for a discussion of the Japanese intervention “exit strategy”.

Official statements in support of more intervention as well as official statements casting doubt on the likelihood of more intervention are both found to be insignificant.

Moreover, the analysis finds strong and robust evidence that for the January 1, 1999 to December 31, 2003 time period the first day of intervention following a no intervention day has a larger than average impact on exchange rates.

Some sample specific evidence that the first day after intervention is associated with an adverse exchange rate adjustment is found.

The analysis does not attempt to address the inherent endogeneity problem concerning intervention studies, thus it is likely that the models are affected by simultaneity leading to an underestimation of the coefficient estimates (see, for example, Dominguez and Frankel 1993b and Neely 2005 for a discussion).

Therefore, the magnitude of the coefficient estimates is not discussed in this paper.
The rest of the paper is organized as follows.
Section 2 details the data.
Section 3 presents the empirical approach of the time-series analysis.
Section 4 discusses the results of the baseline model estimations as well as several extensions and robustness checks.
Section 5 concludes.