معامله آمریکا شمالی و سیاست پولی ایالات متحده

This paper investigates how an increase in the United States Federal Fund rate affects the United States economy and how the effects are transmitted to the Canadian economy using the factor-augmented VAR (FAVAR) approach of Stock and Watson (2005) and Bernanke et al. (2005). A distinguishing feature of our model is the disaggregation of the traded goods sector where imports and exports are disaggregated into 12 and 13 industries, respectively. Extra information is provided on the domestic and international transmission mechanisms between the two countries. The factor-augmented VAR method allows impulse response functions to be generated for all the variables in the data set and so is able to provide a comprehensive description of the domestic and international transmission mechanisms between the United States and Canada

What are the dynamic effects of US monetary policy shocks throughout the US and the Canadian economy? This question is answered by applying factor-augmented VAR methodology — based on the Stock and Watson, 1998 and Stock and Watson, 2002 two-step principal component approach1 — of Bernanke et al. (2005) and Stock and Watson (2005) to a large data set which takes into account the traded goods sector, where exports and imports are disaggregated into 12 and 13 industries, respectively.2 The great appeal of using FAVAR models to study monetary policy transmission mechanisms is that they can, in line with the current practice of monetary authorities, who tend to monitor a large number of economic time series before setting monetary policy, potentially employ thousands of data series in the model. Lagana' and Sgro (2011) have investigated the effect of a rise in the US personal income tax rate on the US and Canada using disaggregated traded good sector data on imports and exports. This paper is an extension of that analysis to monetary policy shocks. Employing FAVAR models has at least two advantages over traditional VARs.3 First, the fact that the VAR information set is likely to be smaller than that used by policy-makers may imply that the VAR model is miss-specified. If the model suffers from omitted variable bias it implies that policy shocks cannot be fully recovered from VAR innovations and so may produce erroneous results. The “price puzzle” discovered by Sims (1992) is only one example of the serious consequences of a miss-specified VAR model.4 Second, impulse responses can only be routinely generated for the variables included in the model, and constitute only a subset of the variables of interest to economists and policy-makers.5 Our approach is to use principal components analysis to calculate the factors that summarize the most relevant information contained in the series. It should be stressed that the number of factors will be much smaller than the number of variables in the data set. As a result, the amount of information which can be handled by the model increases dramatically, hence the chance of under-specifying the econometric model used to assess the effect of policy shocks is significantly reduced.6 The results show that the factor-augmented VAR model generates a reasonable response to the domestic and international transmission mechanisms of United States changes in monetary policy where, following a rise in the United States Federal Fund rate, United States investment and production, inflation and the share price fall. On the other hand, the response of the Canadian economy is mixed. Canadian industrial production, construction and the share price fall, the exchange rate on the US $ to national currency unit (Canadian $) appreciates, while unemployment rises. Imports into Canada from the United States generally increase, while mixed evidence is found on exports from Canada to the United States. Some of these results are in line with standard theory when one considers that the United States is a large open economy whilst Canada is an adjacent, “small”, open economy, and that the trading relations between the two countries are extensive and strong, especially after the signing of NAFTA on 2 January 1988 and its coming into effect on 1 January 1989. However, the extra information generated by the factor-augmented VAR also unearths other identification issues. In particular, following a rise in the US interest rates the US real effective exchange rate depreciates. There are two important points here. First, Bernanke et al. (2005) method in identifying a monetary policy shock using a Cholesky decomposition of the variance covariance matrix is followed. However, as stressed by Bernanke et al. (2005), factor models can be used with a variety of identification methods and techniques designed to overcome identification difficulties.7 Second, these counter-intuitive responses are only evident because of the factor model's ability to generate impulse responses for a large number of variables. Thus, the new “puzzling” responses are evidence of the extra information generated by factor-augmented VARs. This paper is organized as follows. Section 2 describes the econometric framework, Section 3 describes the data, and Section 4 reports the econometric results obtained from the models presented in Section 2.

This paper investigates how an increase in the US Federal Fund rate originating in the United States affects the United States economy and how these effects are transmitted to the Canadian economy using the factor-augmented VAR (FAVAR) approach of Stock and Watson (2005) and Bernanke et al. (2005). A distinguishing feature of our model is the disaggregation of the traded goods sector where imports and exports are disaggregated into 12 and 13 industries, respectively. This provided extra information on the domestic and international transmission mechanisms between the two countries. The factor-augmented VAR method allows impulse response functions to be generated for all the variables in the data set and so is able to provide a comprehensive description of the domestic and international transmission mechanisms of changes in the US Federal Fund rate to a large set of macro variables in the Canadian economy. Following a rise in the US Federal Fund rate, United States unemployment, output, and inflation fall. Canadian employment and output fall, the share price indexes falls, the exchange rate on the US $ to national currency unit (Canadian $) appreciates, some measures of inflation and of the interest rate contract, others rise. Exports to the United States from Canada generally increase, while mixed evidence is found on imports from the United States to Canada. The extra information generated by the factor-augmented VAR also unearths other identification issues. In particular, a rise in the interest rates is associated with a depreciation of the US real effective exchange rate. These are counter-intuitive responses so that the interest rate innovation does not precisely capture a shock to monetary policy. As stated above, there are two important points to note. First, we follow Bernanke et al. (2005) in identifying a monetary policy shock using a Cholesky decomposition of the variance covariance matrix. However, as stressed by Bernanke et al. (2005), factor models can be used with a variety of identification methods and techniques designed to overcome identification difficulties. Second, these counter-intuitive responses are only evident because of the factor model's ability to generate impulse responses for a large number of variables. Thus, these new “puzzling” responses are evidence of the extra information generated by factor-augmented VARs.