بازارهای امنیت و محتوای اطلاعات نقاط عطف سیاست های پولی

Substantial evidence shows that a significant relationship exists between Federal Reserve monetary policy signals and subsequent security returns. Recent evidence, however, suggests that Fed rate changes do not signal shifts in monetary policy and therefore have no real policy significance. In this study, we investigate whether certain Fed signals, characterized as turning points in the monetary cycle, have real policy significance. Our evidence suggests that the Fed's signal that a turning point is occurring is unambiguous, predicts a substantial shift in Fed monetary policy, and provides costless and meaningful information about future security market returns.

There is little doubt that Federal Reserve actions occupy a prominent role in financial market participants’ information sets. The relevance of Fed policy actions is evidenced by the numerous studies that have found significant short-term interest rate and security price responses to policy signals (e.g., Smirlock & Yawitz, 1985; Cook & Hahn, 1988; Wagster, 1993). These short-term reactions are generally characterized as stemming from an “announcement effect.” The identification of an announcement effect associated with Fed policy actions implies that policy signals convey information to the financial markets that is relevant for security pricing. While one would expect that the information conveyed by a Fed policy signal would relate to future monetary policy variables, Thornton (1998a) finds evidence refuting this contention. This evidence raises questions regarding whether signaled changes in Fed policy correspond with actual changes in monetary and reserve aggregates (i.e., are Fed policy signals credible). Several more recent studies suggest that Fed signals also provide information about subsequent long-term security returns (e.g., Jensen & Johnson, 1995; Patelis, 1997 and Thorbecke, 1997; Conover, Jensen, & Johnson, 1999). These studies identify a systematic relationship between long-term stock and bond returns and previous Fed policy changes. Specifically, the studies indicate that security returns are generally higher during periods following an expansive policy signal and are generally lower following a restrictive policy signal. While several studies have confirmed the existence of security return patterns associated with Fed policy signals, the timing of the security return patterns and their relationship with actual changes in monetary policy variables have not been addressed in the literature. In this study, we consider two issues stemming from the research discussed above: (1) Are Fed policy signals credible and (2) What do the security patterns look like over time, and how do they relate to actual changes in monetary policy variables? By considering these two issues jointly, we provide a clearer indication of the relationship between Fed policy signals, monetary policy measures, and security returns. Our study extends previous studies in several ways. First, while Thornton (1998a) evaluates movements in monetary aggregates following each change in the discount rate, we focus only on discount rate changes that signal “turning points” in the monetary cycle. Specifically, we argue that rate changes that are in the opposite direction of the previous change (a turning point rate change) signal that the Fed is shifting its policy stance, while rate changes in the same direction as the previous change simply confirm that the then-current policy stance will continue. Thus, we believe it provides a new and alternative view to focus on “turning point” rate changes to determine whether policy signals convey information about future Fed actions. Second, we evaluate movements in monetary policy variables over both a short-term and a long-term window in order to further assess the relationship between policy signals and subsequent Fed actions. It is conceivable that a decision today by the Fed to adjust its policy stance might lead to observable changes in monetary and reserve aggregates with considerable lag. If this is indeed the case, this could help to explain Thornton's (1998a) findings of no significant short-term movements in the variables before versus after rate changes.1 Third, while several previous studies (e.g., Jensen, Mercer, & Johnson, 1996, Booth & Booth, 1997, and Johnson, Buetow, Jensen, & Reilly, 2003) have documented long-term security return patterns following Fed policy signals, these studies have not examined the timing of the return patterns. Thus, it is unclear whether the return patterns result primarily from security prices moving shortly after the Fed announcement or whether the movements are lagged considerably from the timing of the announcement. Fourth, by examining the return patterns in concert with movements in monetary policy variables, we can determine the degree of correspondence in the two. Specifically, we can determine whether the evidence is consistent with the view that the security return patterns around policy signals result because of Fed induced changes in monetary policy variables. Thus, this is the first study that attempts to provide an explanation for the observed security return patterns around policy changes. Finally, in addition to the three monetary aggregates evaluated by Thornton (1998a), we add the federal funds rate to our analysis. Several researchers advocate the fed funds rate as a superior indicator of actual Fed activities (e.g., Bernanke & Blinder, 1992; Laurent, 1988). Given that the federal funds rate is a market determined rate, it may react more quickly (relative to movements in monetary aggregates) to Fed activities, and thus, provide a better indicator of the Fed's short-term activities around a policy signal. Furthermore, Cook and Hahn (1988) argue that changes in the discount rate have been used to signal the Fed's intention to change its target for the fed funds rate, while Thornton (1998b) indicates that turning-point discount rate changes frequently align fairly closely with turning points in the Fed's federal funds rate target. We believe these studies motivate an empirical examination of the relationship between changes in the fed funds premium and discount rate turning points to determine more precisely the temporal relationship between the two. The remainder of the paper is organized as follows. Section 1 provides a discussion of the sample period, the monetary policy indicators, and the security return indices. In Section 2, we discuss the methodology and report empirical findings. The final section of the paper presents our summary and conclusions.

A substantial body of literature shows that policy signals by the Federal Reserve affect U.S. security returns and interest rates via an announcement effect. It is often argued that interest rate changes by the Fed convey valuable information to the markets, either because they signal new information about economic developments, or because they signal a shift in monetary policy. In related studies, a close association between Fed policy signals and subsequent long-term security returns has been discovered. Whether policy signals convey actual information regarding Fed policy, however, remains an open question in the literature. In this study, we examine whether certain Fed policy signals, specifically those that signal “turning points” in the monetary cycle, provide credible and meaningful information about the subsequent behavior of alternative policy indicators. We also examine the behavior of stock and bond market returns around turning point signals in order to assess the short and long-run value of the information to capital market participants. Our results provide an alternative view of interest rate changes by the Fed as signals of monetary policy shifts. We find that the growth in reserve aggregates (total reserves and non-borrowed reserves), and the trend in the federal funds premium, are significantly altered around these signals. The results strongly support the contention that turning point rate changes credibly predict, or confirm, shifts in Fed monetary policy. The evidence further shows that the adjustment in reserve aggregates occurs gradually over subsequent months. Interestingly, our evidence reveals that an examination of reserve aggregates in the few months surrounding signaled turning points in the monetary cycle could provide a misleading impression of the Fed's actions subsequent to a turning point. Specifically, the overwhelming majority of evidence with respect to the short-term movements in monetary policy indicators suggests that signaled shifts in Fed policy have no policy significance. In contrast, our evidence on the long-term movements in policy indicators strongly supports the policy significance of signaled shifts in Fed policy. Additionally, while changes in the Fed's discount rate and the Fed's federal funds rate target are frequently fairly closely aligned, the federal funds premium displays little movement before a policy shift that is signaled by a discount rate change. After such signals, however, the federal funds premium moves considerably. This result does not support the view of discount rate changes (at least turning point changes) as symbolic and lagging prior changes in the fed funds rate. Therefore, it appears that turning point signals from the discount rate contain valuable information that could not have been captured by viewing prior developments in the fed funds market. Finally, we find movements in short- and long-term security returns that correspond with the changes in the monetary policy indicators. As with the movements in the monetary policy indicators, the security return patterns are shown, for the most part, to develop gradually over time. The patterns in monetary policy indicators and security returns over the 2-year period following a signaled shift in Fed policy are quite similar and are very prominent. The corresponding patterns in monetary policy indicators and security returns is consistent with the contention that the security return patterns exist because of the actions carried out by the Fed following a signaled shift in policy.