پیش بینی های بانک مرکزی و سیاست فاش سازی : چرا برای خوش بین بودن پرداخت انجام می شود؟
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|23158||2007||21 صفحه PDF||سفارش دهید||9976 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : European Journal of Political Economy, Volume 23, Issue 1, March 2007, Pages 30–50
In a model with forward-looking behavior, we study disclosure policy when a central bank has private information on the future state of the economy. We find that the effects of advance disclosure depend on the presence of uncertainty about policy targets when the shock occurs. With uncertainty about policy targets, disclosure is harmless to current outcomes, owing to the strong dependence of inflation expectations on policy actions, which induces the central bank to focus exclusively on price stability. If the central bank's targets are common knowledge, disclosure of future shocks impairs stabilization of current inflation and output.
Over the past several years, many central banks (especially those engaging in inflation targeting) have become more open about releasing their internal forecasts of the state of the economy. In this case, almost all inflation targeting central banks publish their inflation forecasts although some do not publish their output forecasts and nearly all central banks do not announce projections of their interest rate policy path (see Mishkin, 2004 for a detailed discussion). Mishkin asks whether being more transparent helps a central bank “to do its job — that is, enable it to conduct monetary policy optimally with an appropriate focus on long-run objectives?”, and says that “the answer might well be no.” From the private sector's point of view, publication of central bank forecasts is welcome because for some reasons, central bank forecasts outperform those of the private sector, an indication perhaps of central bank's superior information about the future state of the economy, including the state of shocks affecting economic activity. For instance, in their empirical analysis on differences between commercial and Federal Reserve (Fed for short) forecasts, Romer and Romer (2000) conclude that “the most important finding … is that the Federal Reserve appears to possess information about the future state of the economy that is not known to market participants.” (p.455), (emphasis ours). 1 The theoretical literature has explored disclosure of forecasts in the context of a discretionary monetary policy with private information about shocks to current inflation and output. 2 The results are usually mixed and depend on whether there is also additional asymmetry regarding the central bank's preferences. Under a perfectly credible monetary policy, it turns out that rationalizing disclosure of forecasts is usually difficult, at least theoretically ( Gersbach, 2003, Cukierman, 2001 and Jensen, 2000). For instance, in a static setup that features a Lucas-type aggregate supply function, Gersbach (2003) and Cukierman (2001) show that a central bank can improve stabilization policy by withholding its private forecasts of current real shocks. Cukierman (2001) finds a similar result using a simplified version of the backward-looking model of Svensson (1997) that features time lags from the policy instrument to policy goals, and when interest rate variability enters the loss function. Crucial here is that, in both transmission mechanisms, what matters is past expectations of current inflation and the public forms expectations before policy decisions are taken. When there are concerns about the central bank's credibility and private sector expectations respond to central bank actions, the formation of private sector expectations plays a crucial role in determining equilibrium outcomes. Geraats (2001) argues that uncertainty about the inflation target can give rise to credibility concerns. In a two-period framework, Geraats shows that, if the public uses monetary policy actions to infer the unobserved inflation target, there is an incentive for the central bank to invest in reputation in the first period in order to have more flexibility to react to shocks in the second period. This incentive is stronger the more the public knows about current period shocks that the central bank is responding to.3 However, in a two-period New-Keynesian framework that features forward-looking expectations, Jensen (2000) shows how releasing forecasts of current period shocks distorts stabilization policy, even though it solves the credibility problem arising from overambitious (unobserved) output target. Like Geraats, Jensen assumes that the public observes central bank actions before forming inflation expectations, and thus, with a high degree of transparency about current shocks, inflation expectations become extremely sensitive to the central bank's current action. But, here comes the difference, in order to stabilize inflation expectations, policy tilts heavily toward inflation stabilization, making current output very volatile. In that case, transparency could be undesirable for a central banker who enjoys good initial reputation. This paper considers disclosure policy about central bank forecasts and, unlike the previous literature, analyzes the impact of forecasts of future shocks, as these forecasts are of interest when inflation expectations are forward-looking. The model used to analyze future shocks is based on the now familiar New-Keynesian model (see e.g., Clarida et al., 1999, King, 2000 and McCallum and Nelson, 2000), which has forward-looking inflation expectations influencing current period outcomes. In this case, given the central bank's policy, a higher variability of inflation expectations (which are conditional on forecasts of future shocks) means corresponding higher variability of current inflation. This makes disclosure policy regarding forecasts of future shocks an interesting issue to study. We find that the effects of advance disclosure of forecasts of future shocks depends on the presence of uncertainty about the central bank's preference. In line with previous studies, when there is no credibility problem and/or the central bank's preference is common knowledge, disclosure of future shocks impairs stabilization of current inflation and output. On the other hand, when there is uncertainty about the central bank's future preference shock, advance disclosure of future cost-push shocks is harmless to current inflation and output. The reason lies in the strong dependence of one-period-ahead private sector inflation forecasts on central bank actions, which induces the central bank to focus exclusively on price stability. Another implication of the paper is that when withholding forecasts of future shocks, these forecasts may not be revealed to the public by current policy actions, as the central bank prefers to stabilize the effects of private sector forecasts on current inflation. In Section 3, we present a simple benchmark case where the central bank has full credibility and its preferences are common knowledge. In such an environment, the central bank prefers not to release information on future shocks as long as it has other goals besides price stability. With multiple goals, the central bank would like to spread the effects of adverse supply shocks on inflation and output gap, but knowing this, expected movements in future supply shocks make private sector inflation expectations to be more volatile. This effect passes to current prices through expectations of future inflation. Thus, it may be better from the perspective of the central bank not to disclose information until the private sector sets inflation expectations. This ensures that public expectations of future shocks are less volatile than when information about future shocks is available. The benchmark case is then modified in two ways. First, instead of discretionary policy, the central bank is assumed to commit credibly to some state contingent rule (Section 4). This modification does not, however, change the result found under the benchmark case. Second, we introduce uncertainty on the part of the public regarding the central bank's preferences, which gives rise to reputation considerations (Section 5).4 The relevance of shifts in preferences depends on whether such uncertainty is directly resolved or via signaling of policy actions. It turns out that if shifts in the central bank targets have to be inferred indirectly from future central bank actions, disclosing information regarding future shocks can be harmless. Concluding remarks are given in Section 6.
نتیجه گیری انگلیسی
Over the past several years, many CBs have increased their level of transparency with respect to their forecasts on the state of the economy. However, the recent theory on transparency has not settled the question about welfare gains from advance disclosure of CB forecasts. Existing research has analyzed this question assuming private information about current shocks, as these shocks have direct impact on current economic variables, such as inflation and output, that a CB is interested in stabilizing. Based on this notion of private information, a few empirical studies on transparency lend support to the argument that disclosure of CB forecasts can enhance the reputation and flexibility of monetary policy. This paper explores the significance of private information on future shocks, as forecasts of future shocks are crucial when inflation expectations are forward-looking. In a simple macro model with forward-looking expectations, we study disclosure policy when a CB has private information on future shocks. We find that the effects of advance disclosure of forecasts of future shocks depend on the existence of CB preference uncertainty. In line with previous studies, when there is no credibility problem and/or the CB's preference is common knowledge, disclosure of future shocks impairs stabilization of current inflation and output. On the other hand, when there is uncertainty about the CB's future preference shock, advance disclosure of future shocks is harmless to current inflation and output. The reason lies in the strong dependence of one-period-ahead PS inflation forecasts on CB actions, which induces the CB to focus exclusively on price stability. Another implication of the paper is that when withholding forecasts of future shocks, these forecasts may not be revealed to the public by current policy actions, as the CB prefers to respond to PS forecasts. Thus while current shocks may be revealed by current CB actions, this may not be true for forecasts of future shocks. The result about the destabilizing effect of early disclosure of forecasts goes through for some alternative specifications, as long as there is full information regarding CB preferences. The results go through for a loss function that includes interest rate stabilization objective, on top of inflation and output; or if the CB targets nominal income growth, instead of inflation and output, as proposed by some economists. Moreover, whether policy is conducted under discretion or some form of commitment is inconsequential to the main result. Even though disclosing information seems counter-intuitive, as it improves the accuracy of PS inflation forecast, the negative result on welfare (under a credible CB) is a consequence of the CB having objectives other than price stability. With multiple macroeconomic goals, releasing internal forecasts before the public have currently formed expectations of future shocks, and thus future inflation, can actually impair overall stabilization efforts. Our conjecture is that the results also apply if we drop rational expectations and assume in line with the learning literature that the PS and/or the CB adaptively learn about the structure of the economy, adjusting their forecasts with the arrival of new data. All that is needed for our results is that the CB has superior information about future supply shocks. Obviously, there are some limitations of our analysis, limitations that are also shared by the literature on disclosure policy of CB forecasts. First, the CB is assumed to observe PS expectations without error. Introducing observation errors would put the CB at a disadvantage, and with very large errors, the CB may even be forced to be transparent about its private information. Second, there is no strategic manipulation of expectations by the PS, although it knows that the CB is responding to PS expectations. If the PS knows that the CB reacts to PS expectations, there is possible strategic behavior. Third, on the part of the CB there could be a manipulate private information and truthful revelation may not be feasible. As Romer and Romer (2000) have noted, even if CBs disclose their internal forecasts to the public in a timely manner, it is not clear if they would report their true forecasts, or if they adjust them so as to simply follow the markets. Fourth, it would be interesting to extend the analysis so that current and future shocks remain private information of the CB.