今回は、日本銀行の為替介入について、より理解を深めていただける先行研究をご紹介します

結論を一言で述べると、通貨当局は、覆面(秘密での)介入を実施することによる為替レートへの影響と一般的な為替介入政策の方法との一貫性とのトレードオフに直面していることになるのです

今回の投稿から外国為替市場への適切な介入方法(appropriate way to intervene in the FX markets)について考えていただけたのであれば、幸いです💗

This paper has investigated the various steps undertaken by the BoJ when intervening on the YEN/USD market.

We have studied the determinants of three outcomes: (i) the decision to intervene; (ii) the use of secret versus public interventions; and (iii) the detection of secret interventions by FX traders.

The estimation procedure is based on a nested logit specification with three different levels.

The specificity of this model allows us to consider the various outcomes in an integrated framework with a clear distinction between the levels.

Our results allow us to put previous empirical findings on the behavior of central banks in the FX markets into perspective.

On the whole, FX authorities are found to intervene to reduce misalignments and to use their communication policy to further clarify their objectives.

Once the decision to intervene has been taken, the characteristics of the intervention depends on market conditions and the central bank’s overall credibility.

That is, visible operations are used when trying to reverse the prevailing trend, while the failure of previous interventions encourages the central bank to opt for secrecy.

Importantly, a secret intervention has to remain undetected to ensure its consistency with the secrecy strategy.

The way interventions are conducted is then of overwhelming importance in influencing market participants’ detection process.

In this respect, our estimates suggest that a central bank trying to keep its interventions undetected should choose to intervene unilaterally, with low amounts and in a sporadic way.

They also suggest that when the risk of being detected is relatively high, central banks might prefer to intervene in a transparent way to preserve some credibility.

This illustrates that the various decisions relative to the intervention policy are highly interdependent.

Our results also show that a full understanding of any intervention policy has to integrate a large set of variables influencing the three levels discussed above.

For instance, investing low amounts to reduce the probability of detection of an intervention means that the portfolio and microstructure effects are deliberately weakened.

In other words, the central bank lowers its chances of moving the FX trend successfully.
This puts the focus on a potential tradeoff for FX authorities between the necessity of influencing the exchange rate adequately and the consistency with its general intervention strategy.

Importantly, this dilemma only emerges when secrecy is required.

Finally, while our results are specifically related to the BoJ’s intervention policy, we believe that they shed some interesting light on key variables influencing the exchange rate policy decisions of central banks in general.

Most of the results we obtained are consistent with the opinions of a large number of central bankers about the appropriate way to intervene in the FX markets.

Definition 1.
A public intervention is an intervention firmly or clearly reported in newswires (whether through direct detections by market participants or by official speeches confirming the operation).

Here ‘‘clear” means that the report removes any ambiguity concerning the occurrence of the operation and reflects the willingness of the central bank to be detected by the market.

Definition 2.
A secret intervention is an intervention for which there was either no report at all or a report with a high degree of uncertainty.

News reports with a high degree of uncertainty can be easily recognized as they generally contain words such as ‘‘likely”, ‘‘believed”, ‘‘may have/think”, ‘‘rumor”, ‘‘unconfirmed”, ‘‘suspected/ speculation” and ‘‘covert”.

Definition 3.
A pure secret intervention is a secret intervention which is not detected by the market (i.e. for which there were no report of the intervention at all).

Definition 4.
A detected secret intervention is a secret intervention which was detected by the market but with a significant degree of uncertainty (i.e. for which there was at least one uncertain news report about the intervention).