شناسایی شوک های سیاست پولی با تغییر در عملیات بازار آزاد

In this paper, we reexamine the effects of monetary policy shocks by exploiting the information contained in open market operations. A sticky price model is developed where money is the counterpart of securities deposited at the central bank. The model's solution reveals that a rise in central bank holdings of open market securities can be interpreted as a monetary expansion. Estimates of vector autoregressions for US data are further provided showing that reactions to an unanticipated rise in open market securities are consistent with common priors about a monetary expansion, i.e., a decline in the federal funds rate, a rise in output, and inertia in price responses. Compared to federal funds rate shocks, prices do not exhibit a puzzling behavior and a larger fraction of the GDP forecast error variance can be attributed to open market shocks. However, the explanatory power of the latter has decreased since federal funds rate targets have been announced.

Short-run effects of monetary policy have always been of utmost interest for macroeconomists. In the recent past, research in this field has almost reached a consensus with regard to the empirical method by applying vector autoregressions (VARs). Though, different identification schemes have been utilized in the literature (see, e.g., Christiano et al., 1996; Leeper et al., 1996; Sims and Zha, 1998, or Bernanke and Mihov, 1998), the policy measure and some main results for US data are in common in most of these contributions. Monetary policy shocks are usually identified with changes in the operating target, in particular, the federal funds rate,1 while an unanticipated decline of the latter is found to lead to a strong and persistent rise in real activity and inertia in price reactions. This empirical evidence, which is summarized in Christiano et al. (1999), is – at least broadly – consistent with common priors on the impact of monetary policy. Nevertheless, some questions regarding monetary policy effects are still left open. For example, output and, in particular, aggregate prices often exhibit puzzling responses to changes in the federal funds rate (see Sims, 1992; Uhlig, 2001; Hanson, 2002), while simultaneity makes it difficult to isolate exogenous policy shocks from endogenous policy responses.

In this paper, we exploit the informational content of the flow of funds in open market operations to assess the short-run effects of monetary policy shocks. It is shown in a macroeconomic framework that an unanticipated rise in central bank holdings of open market securities can serve as a measure for an expansionary monetary policy shock. Impulse responses computed from fitting VARs for US data do not exhibit a puzzling output or price behavior. Open market shocks are found to contribute to a larger fraction of the GDP forecast error variance than shocks to the federal funds rate or to non-borrowed reserves for the time interval 1959:1 to 2002:2. Impulse responses to GDP innovations further reveal that the open market measure is less sensitive to macroeconomic developments than the federal funds rate, indicating that the balance sheet composition is (almost) exogenously controlled, whereas the federal funds rate is more susceptible to disturbances caused by non-policy factors.