ارزش شفافیت بانک مرکزی زمانی که ماموران در حال یادگیری هستند
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|23159||2007||21 صفحه PDF||سفارش دهید||9755 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : European Journal of Political Economy, Volume 23, Issue 1, March 2007, Pages 9–29
We examine the role of central bank transparency when the private sector is modeled as adaptive learners. In our model, transparent policies enable the private sector to adopt correctly specified models of inflation and output while intransparent policies do not. In the former case, the private sector learns the rational expectations equilibrium while in the latter case it learns a restricted perceptions equilibrium. These possibilities arise regardless of whether the central bank operates under commitment or discretion. We provide conditions under which the policy loss from transparency is lower (higher) than under intransparency, allowing us to assess the value of transparency when agents are learning.
The aim of central bank transparency is to lessen or eliminate informational asymmetries between central bank decision-makers and the private sector. Transparency of central bank decision-making has increased rapidly in recent years beginning with the adoption of inflation targeting by the central banks of New Zealand, Canada, the U.K. and Sweden in the early 1990s.1 Over the same period, economists have made substantial progress in modeling the adaptive process by which learning agents form and update their expectations with the aim of assessing whether rational expectations equilibria can be learned or not.2 However, as Svensson (2003) points out, the connection between central bank transparency and the stability of equilibria under adaptive learning has been largely neglected. Presumably, the benefits of central bank transparency lie in more accurate expectation formation by the private sector, and in improved policy outcomes for central bankers. Nevertheless, a recent literature on the stability of monetary policy rules when the private sector is learning largely ignores the role played by central bank transparency.3 In this literature, the stability of rational expectations equilibria in the benchmark New Keynesian, sticky-price model is assessed under a variety of central bank rules for the interest rate target. The aim is to consider restrictions on the class of policy rules or on policy weighting parameters that ensure stability of rational expectations equilibrium under private sector adaptive learning. The private sector agents in these models are engaged in the process of forming expectations of future inflation and output without regard to the transparency of central bank policy. Consequently, there really is no possibility of assessing the role played by greater transparency on private sector learning behavior and the achievement of central bank objectives. In this paper, we reconsider private sector learning in the context of the New Keynesian model with the aim of understanding the value of central bank transparency. Specifically, we first examine the consequences, for equilibrium stability under learning, of whether or not the central bank reveals its inflation and output targets or that it has committed itself to following a policy rule. Revelation of this information impacts on the specification of the perceived law of motion that agents use to form forecasts of future inflation and output. In the intransparent4 case, where targets are not revealed and/or the private sector is uninformed of the central bank's commitment to a policy rule, the equilibrium is a “restricted perceptions” equilibrium (RPE). In such an equilibrium, agents use a misspecified, underparameterized forecast model of inflation and output but, in equilibrium, they are unable to detect that their model is misspecified. 5 If agents instead used an overparameterized forecast model, they could always learn that the extraneous coefficients in their model should be set to zero and thereby learn to form forecasts consistent with the REE. 6 However, with an underparameterized model that possibility is ruled out; for this reason the underparameterized misspecification and the stability of the resulting RPE is the more interesting case to consider and is the focus of our paper. In the transparent regime, the central bank reveals its policy targets and/or its commitment to a policy rule and, as a consequence, the private sector adopts the correct forecast model. The resulting equilibrium is the standard rational expectations equilibrium. We show that restricted perceptions and rational expectations equilibria are both stable under adaptive learning behavior on the part of the private sector, regardless of whether the central bank operates under a discretionary or a commitment regime. Given this finding, we move on to a comparison of the central bank's policy losses under the two policy regimes. In the discretionary regime, we provide conditions on inflation and output targets under which the policy loss under transparency is lower or higher than under intransparency. Thus under discretion, the case for central bank transparency is mixed. This finding also has implications for the literature on inflationary bias under discretionary monetary policy. In particular, we show that under discretion and intransparency, the central bank may gain from targeting output above the natural rate; the private sector does not incorporate the central bank's policy into its forecast model, and the restricted perceptions equilibrium in which the private sector finds itself does not lead the private sector to question its misspecified forecast model. Finally, in the case where the central bank chooses to commit once-and-for-all to an optimal policy rule, we show that transparency on the part of central bank always leads to a lower loss than does intransparency. Hence, the value of transparency is unambiguously positive when the central bank operates under commitment.
نتیجه گیری انگلیسی
Despite talk by central bankers about transparency of monetary policy goals and targets, vagueness remains en vogue. For instance, in the U.S., the Fed's policy with regard to inflation targeting was recently characterized as follows: “Though Mr. Greenspan adamantly opposes setting any official inflation target, Fed officials have implicitly aimed to keep core inflation near 2 percent.”16 Given this vagueness about monetary policy goals, it would not be so surprising if the private sector assumed that policy targets were zero, leading them to adopt a misspecified perceived law of motion, as we have assumed. Further, as we have shown, the announcement of target levels for output and inflation may not suffice; in the case where the central bank operates under commitment, the public would have to be made aware that their perceived laws of motion should include lagged values of the output gap. In other words, commitment is not enough if agents are learning; you also have to instruct the private sector on how to form their expectations. In this paper, we have made a first effort to link transparency of monetary policy to the specification of the forecast rule adopted by the private sector. We have equated transparency of policy with the private sector's adoption of the correct reduced form forecast model and intransparency with the private sector's adoption of an underparameterized, misspecified model. This view of central bank transparency is quite different from previous notions. It requires that the public shares (or is aware of) the policymaker's model of the economy in the sense of having the same reduced form perceived laws of motion. This stands in contrast to the more traditional view equating transparency with more or better information. We have shown that for certain classes of underparameterized models, the resulting restricted perceptions equilibrium is expectationally stable, meaning that the private sectors' misspecified forecast rules would lead to outcomes for inflation and output that would not invalidate the private sectors' beliefs that their forecast rules were correct. The results we present suggest that in the case of discretionary policy, the value of transparency relative to intransparency of policy is ambiguous, and depends on policy target values. A particularly interesting finding arises when the inflation target is zero but the output target is positive, and the private sector neglects to include a constant term in its forecast models due to intransparency of central bank policy. In this case the central bank benefits from its intransparent policy in that its loss under the resulting RPE is less than its loss in the REE, corresponding to transparent central bank policy. This finding runs counter to the standard inflation-bias outcome when agents have rational expectations and cannot be fooled. By contrast, in the case where the central bank operates under commitment, we find that, for the same calibration studied in the discretionary case, our results unambiguously favor transparent over intransparent policy; the central bank does not gain as much from a commitment regime if the public does not adopt the appropriate forecasting rules. This result on the importance of transparency under commitment is novel; all of the prior literature studying the benefits/costs of central bank transparency is in the context of discretionary policy regimes, where greater transparency reduces the central bank's temptation to create surprise inflation. By studying a commitment regime, we have begun the process of disentangling the transparency problem from problems of time-inconsistency. There are several directions for further research. For example, one could examine the case where the central bank is also learning simultaneously with the private sector. The simplest case would be to imagine that the central bank has the correct perceived law of motion for the interest rate and is learning the appropriate coefficients of this PLM. Alternatively, one might imagine that the central bank has a misspecified law of motion for it, i.e., one that is inconsistent with the optimality condition, and explore the resulting RPE in that case. An example of such an interest rate rule might be a simple Taylor-type instrument rule. Given the tremendous effort that has been expended on increased transparency of central bank policy in recent years it is surprising that so little effort has been expended on modeling precisely how transparency might aid the private sector in forming expectations. We hope that the findings presented in this paper will lead others to think more deeply about this important policy question.