2002 Japan Conference: A Summary of the Papers

This paper explores the effectiveness of official Japanese intervention operations in moving the exchange rate and asks whether intervention might be a useful alternative to conventional monetary policy. The inter-bank interest rate in Japan has been virtually zero since early 1999 and the country recently has experienced very strong base money growth. By contrast, broad money growth has been slow and the level of bank credit has fallen sharply each year since 1996. This raises the specter that Japan may be in a "liquidity trap" so that monetary policy is ineffective in stimulating aggregate demand. Would sterilized (non-monetary) foreign exchange market intervention, by depreciating the exchange rate and inducing export-led growth, be an effective alternative to monetary policy under these circumstances?

Whether sterilized intervention is effective in moving exchange rates is a long-standing open question in the literature on international finance. Among the large industrial countries, this is especially true in Japan because to date empirical studies have not been able to analyze Japanese official intervention data -- the Ministry of Finance (MoF) did not make this available publicly until July 2001¹. The MoF now discloses, with a 1-3 month delay, the day of intervention, the amount of yen intervention (bought and sold by its agent, the Bank of Japan, BoJ), and the currency of intervention.

This new data shows that Japan, among the world's governments, is the single largest actor in the foreign exchange market. Over the April 1991 -- December 2000 period, the BoJ (acting as the agent of the MoF) bought (sold) U.S. dollars on 168 (33) occasions for a cumulative total of $304 billion ($38 billion). This dwarfs all other official intervention in the foreign exchange market. For example, Japanese intervention was greater than U.S. intervention over the same period by a factor of more than 30 and was much greater than Bundesbank intervention operations had been when it was responsible for exchange rate policy in Germany.

An important methodological component of this paper is the application of event study methodology in evaluating the effectiveness of intervention "episodes." Almost all earlier empirical studies on intervention use time-series methodology. However, standard time-series techniques are not well suited to the task, because exchange rates are typically highly volatile on a day-to-day basis while official intervention operations are concentrated in sporadic clusters. Viewed in this light, it is perhaps not surprising that time-series based studies tend not to find strong evidence for a systematic link between exchange rate movements and intervention operations. Instead, the event study approach used in the finance literature seems to fit well so that, in this context, a cluster of intervention operations constitutes a natural candidate for identification as a single event.

We apply the event study methodology to the newly available Japanese data—as well as U.S. intervention data—on intervention operations in the USD/JPY exchange rate market over the April 1, 1991 to December 31, 2000 period. Focusing on daily Japanese and U.S. official intervention operations, we identify separate intervention "episodes" and analyze the subsequent effect on the exchange rate. Using the non-parametric sign test and matched-sample test, we find strong and robust evidence that sterilized intervention systemically affects the exchange rate in the short-run.

What determines a successful event?

Defining each day that the BoJ, the Fed, or both were active in the USD/JPY exchange rate market as a separate event is problematic because intervention frequently comes in "clusters"—several days in a sequence, not necessarily continuous. Therefore we define an event as a period of days with official intervention conducted by the BoJ, the Fed, or both, in the USD/JPY exchange rate market in one direction (in terms of purchases or sales), and possibly including a number of days with no intervention. How many consecutive days of no intervention (the "tranquillity" period) can occur while we still consider the surrounding days of intervention to be part of one and the same event? Given the structure of the intervention data, we choose a "tranquillity" period of five days for our baseline results (and let this number vary in the robustness checks).

The length of the pre-event and post-event periods, respectively, needs to be set long enough to capture a "normal" no-intervention performance of the exchange rate. If the length of the periods is set too long, however, a number of instances of overlap of pre- and post-event windows are created. We apply pre- and post-event window lengths of two, five, ten and fifteen days and find the results robust to either window length. The results with two-day event windows are our baseline, but the paper presents a summary of the results with other event window lengths.

We follow our earlier work (forthcoming in The Economics Journal) and apply three alternative criteria for what constitutes a successful intervention event. The first criterion of success is simply whether the direction of the movement in the exchange rate is the same as the direction in which the central bank was intervening, for example, does the value of the JPY relative to the USD increase after JPY are purchased? This measure of successfulness is referred to as the "direction" criterion. The second criterion defines a successful event as one where intervention is associated with a "smoothing" of the exchange rate movement.

The third criterion distinguishes between "leaning with the wind" and "leaning against the wind" events, conditioning each event on the exchange rate movement of the associated pre-event window. When the "direction" criterion is applied to "leaning against the wind" events only, the resulting measure of success has a clear meaning in terms of reversing the exchange rate trend that prevailed up until the intervention occurred. This particular measure is denoted the "reversal" criteria.

The Results of the Event Study

Focusing first on the two-day pre- and post-event window definitions (and the maximum 5-day "tranquillity" period), we identify 43 intervention events over the 10-year period. Comparing the direction of intervention during the event with the change in the exchange rate over the preceding period (the two-day pre-event window), 34 events appear consistent with a "leaning against the wind" intervention policy and, accordingly, nine events appear in line with "leaning with the wind."

The direction of the change in the exchange rate during the post-event window was consistent with the direction of the associated intervention in 31 events. In other words, 31 of the 43 events were successful according to the "direction" criterion, and by application of the sign test (based on a binominal distribution with the probability of an individual success of 50 percent) randomness is rejected at the 99 percent significance level. Furthermore, 24 of the 34 "leaning against the wind" events were successful according to the "reversal" criterion and, again, the sign test (based on the same 50 percent probability parameter) rejects randomness at 99 percent significance level. While 29 of the 34 "leaning against the wind" events were successful according to the "smoothing" criterion, the sign test of "smoothing" (based on a binominal distribution with the probability of an individual success of 75 percent) cannot reject randomness at any conventional significance level.

Comparing these findings to the sign test results based on window lengths of five days, we find the latter to be stronger. With respect to every criterion of success, randomness was rejected at the 99 percent significance level using the 5-day event window. In short, over a period of 5 days, the success of intervention operations is striking. In particular, it is noteworthy that the "smoothing" criterion is also met for intervention success with the 5-day event window.

As a further test of successfulness according to the "smoothing" criterion, the matched sample test is applied to the 34 "leaning against the wind" events and these were, on average, associated with a reversal of the preceding trend. Splitting up the sample in USD purchases and USD sales, we can strongly reject (at the 95 percent significance level or higher) the null hypothesis of no difference in means in both cases – that is, intervention appears to have had a smoothing effect on exchange rate changes.

To address the concern of a high degree of kurtosis ("fat tails") in the exchange- rate series, we regress our sample of leaning against the wind "matched pair" differences (for both USD sales and USD purchases) on a constant term using White's (1980) heteroskedasticity-consistent (robust) standard errors. With rejection at the 99 percent significance level, these regressions are fully consistent with the matched sample tests based on the normal distribution.

Conclusion

We find support for short-run effectiveness of intervention. Our horizon for the baseline results were for an effect measured from 2-5 days; an extension of the framework showed effects lasting for up to two weeks. Intervention also is effective in the short-term even if not accompanied by supporting interest rate change and regardless of whether intervention is "secret" (in the sense of no official reports or rumors of intervention reported over the newswires). This suggests that the Bank of Japan could indeed engineer exchange rate depreciation (thereby counteracting deflation and recession) even though interest rates cannot be moved downwards further. An important caveat, however, is that intervention in our study only appears effective in moving the exchange rate over a short horizon (up to two weeks), clearly limiting its usefulness as a substitute for an effective monetary policy instrument.

1. The only paper using the newly available Japanese intervention data that we are aware of (March 2002) is a study by Ito (2002).

Fatum, Rasmus and Michael M. Hutchison: "Is Sterilized Foreign Exchange Intervention Effective After All? An Event Study Approach", UCSC Working Paper No. 02-02 (August 2001). Forthcoming, The Economic Journal (April 2003).

Ito, Takatoshi: "Is Foreign Exchange Intervention Effective? The Japanese Experience in the 1990s," NBER Working Paper No. 8914, April 2002.