صحت و سقم گزارش های مطبوعاتی در مورد مداخله ارزی بانک مرکزی ژاپن

This paper presents evidence on the accuracy of press reports regarding the foreign exchange market interventions conducted by the Bank of Japan (BoJ) between January 1995 and December 1999. We find that the reports of interventions in the financial press are a relatively inaccurate indicator for the actual interventions of the BoJ. We also find that the accuracy of press reports of BoJ interventions is higher for those interventions that were carried out jointly by the BoJ and the Federal Reserve.

A substantial body of research in the international finance literature has analyzed the effects and the effectiveness of foreign exchange market interventions of central banks. This research has not only broadened our knowledge of how central bank interventions affect the level and the volatility of exchange rates but has, thereby, also yielded important insights into how foreign exchange markets work. A prerequisite for undertaking such research is the availability of high-quality intervention data. With respect to the largest economies and the most important currencies worldwide, intervention data, until recently, were made available to researchers only by the Federal Reserve (Fed) and the Deutsche Bundesbank. In contrast, the Bank of Japan (BoJ) did not release official data on its foreign exchange market interventions.1 As a consequence, researchers who tried to analyze the interventions of the BoJ had to use intervention reports in the financial press to identify BoJ intervention days without knowing how accurate these intervention reports were (see Dominguez, 1998 and Ramaswamy and Samiei, 2000). In the academic literature, there is remarkably little empirical evidence on the accuracy of press reports of central bank interventions. Empirical studies addressing this issue are only available for press reports of Bundesbank and Fed interventions (see Dominguez and Frankel, 1993, Klein, 1993 and Osterberg and Wetmore-Humes, 1993). In this paper, we go beyond earlier studies of the accuracy of intervention reports in the financial press by assessing the accuracy of press reports of BoJ foreign exchange market interventions. To this end, we use the fact that the BoJ recently changed its information policy and released a comprehensive dataset on its interventions in foreign exchange markets.2 This dataset has not yet been used to study the accuracy of press reports of BoJ interventions. Comparing the data on actual BoJ interventions with reports of BoJ interventions published in the financial press during a period from 1995 to 1999, we find that press reports are a relatively inaccurate indicator of the actual interventions of the BoJ. Furthermore, our results suggest that the accuracy of press reports of BoJ interventions is higher for interventions that the BoJ carried out jointly with the Fed. We organize the remainder of the paper as follows. In Section 2, we use our new dataset to provide some descriptive statistics of the actual BoJ and Fed foreign exchange market interventions during the period 1995–1999. This is the period for which we assess the accuracy of reports of BoJ interventions published in the financial press. In Section 3, we analyze in detail the link between the actual and the reported BoJ interventions in the foreign exchange market. In Section 4, we supplement this analysis by estimating a statistical model to test for systematic differences between the actual and the reported BoJ interventions. In Section 5, we offer some concluding remarks.

In this paper, we analyzed whether the reports published in the financial press of BoJ interventions in the yen/US dollar market are sufficiently accurate for their use in empirical studies as a proxy for the actual intervention policy of the BoJ to be reliable. In the intervention literature, press reports were, for example, applied as proxies in empirical studies on the effectiveness of the intervention policy of the BoJ. The use of proxies for BoJ interventions was necessary because, until recently, the BoJ in general did not publish any data on its intervention policy. Because the BoJ recently released a comprehensive dataset on its foreign exchange market interventions, it is now possible to study the question of the accuracy of press reports of BoJ interventions by comparing them with the actual intervention activity of the BoJ. The main message of this paper is that one should be somewhat reserved and cautious when using press reports as proxies of intervention activity in analyses of the intervention policy of the BoJ. Our results indicate that press reports, in fact, provided only a rather crude impression of the actual intervention policy of the BoJ. Although we can confirm that the reports in the financial press contained some information regarding the foreign exchange market interventions of the BoJ, we find that: • the majority of the interventions conducted by the BoJ during the time period 1995–1999 were not reported by the financial press; • the financial press frequently reported BoJ intervention activities although no intervention had actually taken place. Therefore, the recently released official dataset of BoJ interventions rather than the proxy variables for BoJ interventions commonly used in the empirical literature should be used in future analyses of the intervention policy of the BoJ. Given the inaccuracy of the press reports of BoJ interventions we documented in this paper, it may also be worthwhile to use the official BoJ intervention dataset to reassess the results of earlier empirical studies of the foreign exchange market interventions of the BoJ.