نرخ تبدیل ارز و قواعد سیاست پولی بهینه در اقتصاد باز و در حال توسعه : برخی از تجزیه و تحلیل های ساده

Using a simple, tractable model, this paper revisits and expands upon issues relating to optimal monetary policy rules (MPRs) in open economies. The optimality of the rule is explored through various specifications of a central bank loss function as it is the loss function that offers insight into central bank preferences. Many of the issues on this topic have centred on the role of the exchange rate: Is it optimal for the policy instrument to react to the exchange rate? What is the role of the exchange rate in a domestic inflation targeting vs CPI inflation targeting? Does a fear of floating have any bearing on the way optimal MPRs are constructed? While this paper is not empirical, the analysis is relevant for central banks in open and developing economies that face a choice between allowing exchange rates to float (and adopting an inflation targeting regime) and engaging in some degree of exchange rate fixity.

Much of the recent discussion on optimal monetary policy rules has been conducted in the context of the rule being employed as part of an inflation targeting regime (see Ball, 1999, Ball, 2001, Svensson, 2000, Taylor, 2000a, Taylor, 2000b, Cavoli and Rajan, 2007 and Collins and Siklos, 2004 and many others). This is essentially because inflation targeting has become a seriously contemplated policy option for many open economies and recently, for emerging market economies in Southeast Asia (see Carare and Stone, 2003, Masson et al., 1997, Mishkin, 2000, Ito and Hayashi, 2004, Ho and McCauley, 2003 and Cavoli and Rajan, 2006a). But, is inflation targeting suitable for open and developing economies? Given the size and degree of openness of many emerging market countries, the exchange rate may play an important role in the transmission of foreign shocks into domestic economic conditions and also in the transmission of policy actions. As such, there is an added incentive for the exchange rate to be included as part of policy and the resulting multiplicity of objectives risks being at odds with the inflation targeting regime. This paper examines the mechanics of this very problem—the dilemma associated with having an inflation target and an exchange rate objective and one policy instrument. Of central importance in this paper is the role of the exchange rate in the construction of optimal monetary policy rules (MPRs) in an open economy. Optimality in this paper is given by particular specifications of the central bank loss function. An optimal rule, therefore, is one that is derived as the result of minimising a loss function where the variables contained in the function and their corresponding weights reflect central bank preferences. The rule itself, once derived from the central bank minimisation problem, specifies how the instrument of policy reacts to the variables specified in the model. The monetary policy rules derived here differ from those derived under the usual inflation targeting regime in that this paper considers the possibility that the policymakers are interested in a degree of management of the exchange rate. The analysis in this paper pre-supposes the following context: A developing economy central bank has implemented procedures supporting inflation targeting arrangements. However, there remain nagging doubts about committing wholeheartedly to inflation targeting due to concerns about the adverse effects of excessive exchange rate volatility (a fear of floating, see Calvo and Reinhart, 2002). Under the analytical (and, indeed, institutional) framework set up for inflation targeting, can the exchange rate play a role in the implementation of policy? This scenario is not purely conceptual—it has both positive and normative implications that reflect actual policy dilemmas. On the positive side, this scenario is very reflective of the situation in the Asian economies of Korea, Thailand, Indonesia and the Philippines. Recent work has examined the post-crisis de facto exchange rate regimes of these countries and has found evidence of a possible desire to revert to pre-crisis USD or basket pegs (see Baig, 2001, Cavoli and Rajan, 2006b and Willet et al., 2005). McCauley (2001) and Parrado (2004), for instance, examine the role of the exchange rate in Singapore’s monetary policy regime. On the normative side, there is a strong view that inflation targeting systems can be augmented to characterise optimal policy with a degree of involvement of the exchange rate (see Goldstein, 2002, Edwards, 2002, Fischer, 2001 and Chang and Velasco, 2000). For instance, Chang and Velasco (2000) state: “In short, the evaluation of exchange rate policy should move away from the ‘fix vs flex’ dichotomy, and toward the characterisation of optimal monetary policy in well specified analytical frameworks.” (p75). Moreover, Edwards (2002) states: “Indeed, it is perfectly possible that the optimal policy, that is, policy that minimizes a well-defined loss function is one where the central bank intervenes from time to time.” (p248). 1 There is a large literature on the role of the exchange rate with monetary policy rules and it covers a range of contributions—from those that are quite technical (Svensson, 2000, Gali and Monacelli, 2005 and Devereux et al., 2005 among others) to those that are essentially descriptive and policy-oriented (Eichengreen, 2001, Taylor, 2001, Ito and Hayashi, 2004 and Ho and McCauley, 2003). The more technical literature presents, for the most part, simulation/numerical methods as a way to analyse the effectiveness of various policies. This paper attempts to bridge the gap between this literature and the descriptive literature by examining the simple analytics of alternative optimal rules. To my knowledge, there are few papers of this type pertaining to open economies. Clarida, Gali and Gertler (2002), Guender (2005) and Semmler and Zhang (2007) are such examples. The intention of the paper is to present the type of analysis for open economies as Ball (1997) and Cecchetti (2000) conducted for closed economies. While the simple analytics are the focus of the paper, in order to augment the analytics and to illustrate some of the main points, a simple numerical example is presented later in the paper. The paper is structured as follows: The following section presents the basic model and derives the optimal rule for domestic inflation targeting, CPI inflation targeting and real exchange rate targeting. Section 3 presents the interesting policy implication of fear of floating and discusses whether there are instances where fear of floating is justified. Section 4 presents a simple numerical exercise as a way to illustrate some of the results gained in the previous sections. Section 5 concludes.

The experiences of developing economies of Asia and elsewhere suggest that authorities appear to grapple with the need to manage exchange rate stability while attempting to support a policy regime supporting domestic objective (inflation targeting). Much of the literature on exchange rate regimes tend to focus on the polar views of fully fixed or floating exchange rate or the literature focuses on MPRs but is based on numerical analysis and simulation or is estimated empirically. This paper moves away from the fixed vs floating dichotomy and examines the role of central bank preferences in the construction of optimal MPRs. The model and solutions are simple and it is this simplicity that allows the examination of each optimal rule and enables us to see how policy preferences affect the behaviour of the rules. This presents us with the opportunity to see how the rules actually work when policy preferences change (such as from CPI inflation targeting to real exchange rate targeting) or when different scenarios are examined (openness or LD). One of the main themes to emerge from the analytics and the simple numerical exercise is that the interaction between policy parameters can dramatically change the way the optimal rule behaves. This is best seen in the case real exchange rate targeting where, as the importance (to the central bank) of the exchange rate increases relative to output and inflation, the reaction of the policy instrument to current exchange rate increases. The other main theme is that fear of floating behaviour by central banks—that is, central bank preferences geared towards managing exchange rate movements—can be justified under certain conditions. The conditions examined here relate mainly to openness to exchange rate changes and financial vulnerability due to Liability Dollarization. For the case of greater pass-through and greater vulnerability due to LD, the reaction to the current exchange rate is greater than for other variables when compared to the baseline case of no pass-though or LD effect. The results presented here correspond closely to the simulation and empirical literature on this topic in that those scenarios that potentially subject an economy to greater exchange rate exposure will induce an optimal policy response designed to mitigate that exposure.