تورم خروجی تجارت کردن در یک چارچوب اقتصادسنجی مبتنی بر بهینه سازی های اعمال شده بر یک اقتصاد باز : مورد ژاپن

Applying an optimization-based econometric framework to an open economy for the evaluation of monetary policy, this paper examines the dissolving of inflation–output trade-offs, or improving social welfare, through inflation targeting by using data for Japan. We simulate a counterfactual inflation targeting policy with a model following New Keynesian Open-economy Macroeconomics, by estimating the historical feedback rule and deep parameters, and identifying shock processes. The findings indicate that consumer price index inflation targeting with commitment improves social welfare when compared with the historical policy in Japan. It does so through stabilizing not only the producer price inflation rate, but also output.

In the field of monetary policy in the open economy, monetary policy rules, including inflation targeting (IT), are discussed briskly. Concerning the IT, the implications derived by Aoki (2001), who finds that domestic IT dissolves inflation–output trade-offs under a closed economy, are verified by some work in New Keynesian Open-economy Macroeconomics (NOEM). From the NOEM perspective, there are some representative studies that examine the IT. With a small open economy, Gali and Monacelli (2005) show that inflation–output trade-offs are dissolved by Producer Price Index (PPI) IT while Benigno and Benigno (2006) and Okano (2007) show that PPI or Consumer Price Index (CPI) IT stabilize both inflation and output simultaneously within a two-country model. Moreover, Benigno and Benigno (2006) and Gali and Monacelli (2005) assert that this policy improves social welfare, which is derived with a second-order Taylor approximated utility function. The theoretical implications derived by Benigno and Benigno (2006) and Gali and Monacelli (2005) have been recently proven by Okano (Forthcoming), who investigated IT in the UK using the optimization-based econometric framework (OEF) proposed by Rotemberg and Woodford (1997). Extending this framework to the small open economy, Okano (Forthcoming) simulates the UK economy on the assumption that the Bank of England did not adopt IT in 1992:4. The results indicate that not only the simulated PPI inflation rate but also the simulated output is more volatile than both historical paths, being the result of IT. The results also indicate that IT has improved social welfare in the UK when compared with the simulations. At this time, the policy implications derived by Benigno and Benigno (2006) and Gali and Monacelli (2005) are empirically consistent. Following papers assuming an open economy, this paper examines dissolving inflation–output trade-offs, or improving social welfare through IT, using data for Japan after the 1990s. The assumption of an open economy has some importance for the analysis of monetary policy. For example, Aoki (2001) argues that one interesting question is whether the central bank (CB) should stabilize a broad inflation measure, which includes the prices of imported goods, or just domestic inflation. This question may be addressed only from an open-economy perspective. In addition, a more important suggestion is provided by Tille (2001). Taking notice of the Marshall–Lerner–Robinson condition, Tille (2001) shows that taking a closed economy as an approximation of a large open economy can be misleading when domestic and imported goods are poor substitutes. In such an economy, even if the home country is large enough to be regarded as a closed economy, a beggar-thyself effect arises, which depends on the elasticity of substitution between home and foreign goods. The elasticity of substitution between home and foreign goods has a significant effect on macroeconomic fluctuations; however, as long as a closed economy is assumed, it disappears. In this analysis, we adopt the OEF proposed by Rotemberg and Woodford (1997). This framework is employed not only by Okano (Forthcoming) but also by Boivin and Giannoni (2003), Giannoni and Woodford (2003) and Amato and Laubach (2000). According to Amato and Laubach (2000), the essence of this framework is obtaining estimates of structural parameters by minimum distance estimation, and identifying shock processes to replicate the remaining time series features of the endogenous variables. These procedures make it possible to simulate counterfactual monetary policies. The framework consists of two models, including VAR and the microfounded structural model. However, unlike VAR or S-VAR, the framework can explain the changes in effects of monetary policy and can also be used for welfare analysis. While much work using this framework implicitly assumes a closed economy, we employ the OEF applied to an open economy to verify the implications derived by Benigno and Benigno (2006) and Gali and Monacelli (2005). As discussed, studying monetary policy in an open economy is of some importance. We especially focus on the form of inflation measure; i.e., which form is suitable from the viewpoint of dissolving inflation–output trade-offs or improving social welfare. By assuming an open economy, we can address the question of whether the CB should stabilize a broad inflation measure, which includes the prices of imported goods, or merely domestic inflation, as in Aoki (2001). Whereas the OEF is a powerful tool for analyzing monetary policy, a number of additional procedures are required to simulate counterfactual monetary policy. To avoid confusion, our strategy is introduced first. First, Gali and Monacelli's (2005) small open-economy model, with some modification, is constructed. Second, this model is log-linearized. Third, a VAR system including the interest rate feedback rule is estimated. Fourth, structural parameters are estimated and the shock processes are identified. Fifth, we simulate counterfactual monetary policy, namely, IT. Lastly, we compute the social loss from a function derived from a second-order approximation of the utility function. Note that while this framework needs a number of procedures, our aim is to simulate counterfactual monetary policy and compute the social loss. In this regard, while estimating the feedback rule and the structural parameters and identifying the shock processes are important, they are not the principal purpose of the analysis. Our results, which are obtained by simulating the PPI IT and the CPI IT with both discretion and commitment, are summarized as follows. First, it is shown that IT can dissolve inflation–output trade-offs, and there is a tendency for CPI IT to steady the PPI inflation rate and output more than PPI IT. Second, committed monetary policy can reduce the social loss more than by discretionary policy. In particular, CPI IT with commitment is the best policy among four simulated policies from the viewpoint of improving social loss. These implications are consistent with Benigno and Benigno (2006), who point out that the open-economy optimal policy rule generally reacts not only to the PPI inflation rate but also to the terms of trade (TOT). This can be understood by paying attention to the CPI inflation rate, which is decomposed into the PPI inflation rate and the TOT depreciation. Third, we identify CPI IT as one policy prescription for avoiding Japan's great stagnation, thereby verifying Krugman, 1998 and Krugman, 2000. The paper is structured as follows. Section 2 briefly describes the model. Section 3 log-linearizes this model. Section 4 estimates the VAR system including the feedback rule and deep parameters of the structural model and identifies the shock processes. Section 5 simulates the counterfactual monetary policy. Section 6 evaluates counterfactual monetary policy. Section 7 concludes the paper.

There are a number of important relationships between the analysis in this paper and earlier work. With more severe parameter restrictions, Gali and Monacelli (2005) prove that PPI IT can dissolve inflation–output trade-offs in the small open economy. We relax these restrictions and show empirically that CPI IT is more desirable from the viewpoint not only of dissolving inflation–output trade-offs but also of improving social welfare in an open economy; for example, in a country such as Japan. This is our main finding. Using a model closely related to that of Gali and Monacelli (2005), Okano (Forthcoming) shows that IT employed in the UK, where the inflation rate of retail price index excluding mortgage interest payments, which is similar to CPI, is specified as a targeted inflation rate, can improve social welfare. Benigno and Benigno (2006) point out that if the degree of relative risk aversion multiplied by the elasticity of substitution between the home and foreign goods is not unity, the optimal instrument rule reacts to not only the PPI inflation rate but also the TOT. Okano (2007) also shows that the CPI inflation rate is a desirable target in an economy with pricing-to-market because pricing-to-market behavior has relevance to the substitutability between domestic and foreign goods. To that end, this analysis yields similar results. Taking into account the implications of these preceding papers, our main finding then has some validity. Jung et al. (2005) demonstrate that the liquidity trap can be avoided by committing to a zero-interest rate policy. They show that this is desirable monetary policy under the zero-bound nominal interest rate constraint faced by Japan. As an alternative, although we allow for high variability in the nominal interest rate, we illuminate the problem of choice of targeted inflation rate not examined in Jung et al. (2005). The results of our study are also similar to those in Okano (2007), though pricing-to-market is not assumed. However, this paper does not explain why the implications are similar to those derived without pricing-to-market. Future research using OEFs applied to an open economy should take into account the unresolved issues of CPI IT under a zero-bound nominal interest rate and pricing-to-market.