تاثیر افتراقی اطلاعیه های سیاست های پولی و تبیین منتشر شده دقیقه ای در بازار معاملات آتی نرخ بهره در استرالیا

Unlike US and Continental European jurisdictions, Australian monetary policy announcements are not followed promptly by projections materials or comprehensive summaries that explain the decision process. This information is disclosed 2 weeks later when the explanatory minutes of the Reserve Bank board meeting are released. This paper is the first study to exploit the features of the Australian monetary policy environment in order to examine the differential impact of monetary policy announcements and explanatory statements on the Australian interest rate futures market. We find that both monetary policy announcements and explanatory minutes releases have a significant impact on the implied yield and volatility of Australian interest rate futures contracts. When the differential impact of these announcements is examined using the full sample, no statistically significant difference is found. However, when the sample is partitioned based on stable periods and the Global Financial Crisis, a differential impact is evident. Further, contrary to the findings of Kim and Nguyen (2008), Lu et al. (2009), and Smales (2012a), the response along the yield curve, is found to be indifferent between the short and medium terms.

In the US and Continental European jurisdictions, monetary policy announcements that disclose a target interest rate for overnight loans are followed within a period of less than 2 h by statements that explain the decision behind those announcements. For example, the announcement of the setting of the Federal Funds rate by the release of the US Federal Open Market Committee statement is followed 90 min later by the release of so-called “projections materials”. Similarly the European Central bank announces its rates decision and 45 min later releases a comprehensive summary of its stance on monetary policy. In Australia, these events occur separately, with the announcement of the monthly target cash rate for overnight loans being provided 2 weeks before the release of the explanatory minutes of the Board meeting that decided the rate outcome. This paper exploits this feature of the Australian monetary policy environment to examine the differential impact of monetary policy announcements and explanatory statements on the Australian interest rate futures market.1 The impact of Australian monetary policy announcements has been examined. For example, Kim and Nguyen (2008) examined the impact of the Reserve Bank of Australia and US Federal Reserve target rate announcements on Australian financial markets, reporting a stronger (weaker) impact on short (long) term contracts based upon the magnitude of the coefficients in the regression model employed. Lu et al. (2009) confirm these findings using intraday data.2 More recently Smales (2012a), using the methodology of Guraynak et al. (2005), sought to disaggregate the impact of the monetary policy announcement into target and path surprise factors by employing a two-factor regression model.3 However, no Australian study has addressed the impact of the explanatory minutes. International studies have examined the impact of the explanatory minutes in England (Reeves and Sawicki, 2007) and in the US (Jubinski and Tomljanovich, 2013 and Rosa, 2013) by comparing the volatility and/or returns on announcement days with non-announcement days. These studies use a non-market-based measure for surprise such as survey data.4 Further, no examination has been undertaken as to whether there is any differential impact between monetary policy announcements and explanatory minute releases. This study is motivated by the segregation of the Australian monetary policy announcement and explanatory minutes release. This segregation allows the calculation of the information content of these events to be separately estimated and for their impacts to be independently assessed. It also allows for their differential impacts to be examined: in general; and along the yield curve. A wide literature also demonstrates the influence of economic states on the impact of macroeconomic announcements. For example McQueen and Roley (1993) demonstrate that allowing for different stages of the business cycle influences the impact of several macroeconomic announcements.5 Whilst no study has considered the differential impact of monetary policy announcements and the release of explanatory statements across different economic states, this is of particular interest to policy makers, who need to ensure that they deliver effective direction during periods of crisis, when monetary policy is particularly important. We therefore consider the influence of the Global Financial Crisis on the impact of monetary policy announcements and explanatory minutes releases on the Australian interest rate futures market. Using a sample period from October 2003 to July 2012 we find that both monetary policy announcements and explanatory minutes releases have a significant impact on the implied yield and volatility of Australian interest rate futures contracts. When the differential impact of these announcements is examined using the full sample, no significant difference is found. However, when the sample is partitioned based on the Global Financial Crisis, a differential impact is exposed. Further, contrary to the findings of Kim and Nguyen (2008), Lu et al. (2009), and Smales (2012a), the response along the yield curve, is found to be indifferent between the short and medium terms. The paper proceeds as follows. Section 2 presents the data and research design. Section 3 presents the results. Section 4 summarises the paper.

This section is divided into two parts. First, the impact of unexpected monetary policy announcements and explanatory minutes releases on Australian interest rate futures implied yields is examined in Section 3.1. Secondly, in Section 3.2 we test whether there is a differentiable impact between these two events. 3.1. Response to unexpected monetary policy announcements and explanatory minutes releases 3.1.1. Impact on implied yields It is evident from the results presented in Table 1 that the impact of unexpected monetary policy announcements and explanatory minutes releases on Australian interest rate futures implied yields is immediate (within 10 min) and significant. Further, consistent with extant literature (Kim and Nguyen, 2008, Lu et al., 2009 and Smales, 2012a) the magnitude of the impact tends to decline with contract maturity. Specifically, an unexpected 100 basis point tightening of the target cash rate leads to a 0.943%, 0.682% and 0.300% increase in the implied yield of the 90-day bank bill, 3-year and 10-year treasury bond futures contracts respectively. As the RBA generally increases or decreases the cash rate in 25 basis point increments, this result may be contextualised as an unexpected 25 basis point tightening prompting a 0.236%, 0.170% and 0.075% increase in the implied yields of the 90-day bank bill, 3-year and 10-year treasury bond futures contracts respectively. Similarly, an unexpected 100 basis point increase, derived from the information content of the explanatory minutes, leads to a 1.201%, 0.945% and 0.405% increase in the implied yields of the 90-day bank bill, 3-year and 10-year treasury bond futures contracts respectively. These results suggest that the unexpected information content of an explanatory minutes release elicits a larger response from the Australian interest rate futures market than an unexpected monetary policy announcement. This finding is confirmed by examining the mean response of the 30-day interbank cash rate futures contract implied yield to the monetary policy announcement (0.053% change in implied yield) and explanatory minutes release (0.085% change in implied yield). 3.1.2. Impact on volatility Having identified that the response to monetary policy announcements and explanatory minutes releases occurs within the first 10 min, the mean equation of the EGARCH model can be accurately specified, as reported in Table 2. These results indicate a significant volatility response to monetary policy announcements and explanatory minutes releases. The magnitude of this response (like our mean results) tends to decline with contract maturity as denoted by βε1, whilst the persistence of the response, denoted by the GARCH term βσ, tends to increase with maturity. Further, the significance and sign of βε2 denotes that the volatility response is asymmetric, stronger to negative innovations (falling yields) than positive innovations (rising yields). This asymmetry however is only found in the 3-year and 10-year treasury bond futures contracts. The absence of significance in this variable, using the 90-day bank bill futures contract, indicates that the response to these announcements/releases, in the short-term, is symmetrical. The timing of the volatility response, as indicated by the sign and significance of the dummy variable dt, is shown to vary between the two events. Specifically, volatility rises in the 10 min prior to the monetary policy announcement, remains elevated for up to 20 min, then drops significantly, returning to pre-announcement levels. Whilst volatility surrounding the explanatory minutes release, does not peak until announcement time (t0,9) and only gradually returns to pre-release levels, taking up to 40 min. 3.2. Differential impact of unexpected monetary policy announcements and explanatory minutes releases 3.2.1. Differential impact of unexpected monetary policy announcements and explanatory minutes releases The differential impact of monetary policy announcements and explanatory minutes releases on Australian interest rate futures implied yields, previously only inferred by the size of the coefficients (see Section 3.1.1), is reported in Table 3. Consistent with the results reported by Smales (2012a) in his analysis of monetary policy announcements, the intercept term is negative and statistically significant for all three futures rate contracts. Whilst these negative intercept terms indicate that implied yields on interest rate futures contracts on average decrease at the time of monetary policy announcements, their small magnitude, between 0.2 and 0.4 basis points, suggests that this decrease could not be economically exploited. The significant positive coefficients on the intercept dummies (β2) indicate that the release of explanatory minutes does not elicit the same systematically negative response in implied yields. In conclusion, when the full sample is used there is no significant difference between the market's reaction to monetary policy announcements and explanatory minutes releases. The coefficient on the slope dummy (β3) is not significantly different from zero in all three models. 3.2.2. Differential impact of crisis and non-crisis periods Although there was no differential impact between monetary policy announcements and explanatory minutes releases on interest rate futures implied yields when the full sample was used (see Section 3.2.1), partitioning the sample based on the Global Financial Crisis demonstrates differential impacts. While no difference is reported in non-crisis periods (see Panel A), the results reported in Panel B of Table 4 show that during the Global Financial Crisis, an unanticipated 100 basis point tightening of the target interbank cash rate leads to an increase of 1.160%, 0.604% and 0.306% in the implied yields of the 90-day bank bill, 3-year and 10-year treasury bond futures contracts respectively. However, the impact of explanatory minutes releases on interest rate futures yields is significantly reduced. This is denoted by the statistically significant negative coefficients for β3 in all three regressions reported in Panel B. These results indicate that during periods of crisis subtler monetary policy mechanisms, such as the release of the explanatory minutes, are not as effective as direct actions, such as the monetary policy announcement. 3.2.3. Differential impact along the yield curve The differential impact of monetary policy announcements and explanatory minutes releases along the yield curve of Australian interest rate futures contracts is reported in Table 5. It is evident from the results reported in Panel A that the response of 10-year treasury bond implied yields to unexpected monetary policy announcements and explanatory minutes releases is significantly weaker than the response of 90-day bank bill and 3-year treasury bond implied yields. This is indicated by the positive and significant coefficients of β4 and β5 that represent the differential response of these yields (respectively) when compared to 10-year treasury bond implied yields. Specifically, the response of 90-day bank bill (3-year treasury bond) implied yields to an unexpected 100 basis point tightening of the target cash rate is 0.643% (0.382%) larger than the response of 10-year treasury bond implied yields (0.300%). Similarly, the response of 90-day bank bill (3-year treasury bond) implied yields to an unexpected 100 basis point increase, derived from the information content of the explanatory minutes, is 0.796% (0.540%) larger than the response of 10-year treasury bond implied yields (0.405%). In Panel B however the difference between 90-day bank bill and 3-year treasury bond implied yields is insignificant (β4), which indicates that the responses of these two contracts yields are not statistically different. This result is contrary to existing findings ( Kim and Nguyen, 2008, Lu et al., 2009 and Smales, 2012a) regarding the impact of monetary policy announcements along the yield curve.