Ars.els-cdn.com
APPENDIX – VARIABLES EMPLOYEDChinaReal economyFinancialMonetary PolicyConsumer price index90 day interest rate (PBOC)Exchange rate (nominal/real vis-à-vis US dollar)Recession indicatorForeign exchange reserves (US dollars)Discount rate (PBOC)Consumer confidence indexDomestic sovereign debt24 hr central bank rate (PBOC)ExportsTotal credit: households/non-financial firmsM2ImportsStock market index (Shanghai/Shenzhen)Required reserve ratioReal GDP growth (gaps)Property pricesMonetary baseEarnings indicatorConstruction activity indicatorEnergy production indicatorEconomic policy uncertaintyInflation forecast (one year ahead)Real GDP growth forecast (one year ahead)Oil prices (international)Current account balance (to GDP)Foreign direct investmentIndustrial productionCommodity pricesBusiness climate indexNote: In bold the series used to derive factors (see Figure 3).USReal economyFinancialMonetary PolicyPCE deflator3 month commercial paper rateFed funds rateReal GDP growth Total creditExcess reserves of the banking systemEconomic Policy UncertaintyFinancial conditions indexCentral bank total assets (US Fed)Inflation forecast (one year ahead)Total commercial loans (volume)Shadow fed funds rate (Wu-Xia version)Real GDP growth forecast (one year ahead)Long-term interest rate (US govt, 10 years and over) less3-month Treasury bill rate (SPREAD)National Financial Conditions index*Senior Loan Officer Survey indicatorsOil pricesVIXOutput gap (CBO, HP)Housing pricesUnemployment rate gap (relative to ‘natural’ rate)Stock market index (S&P, Wilshire)NBER recession datesDebt/GDP ratioPrivate non-residential fixed investmentNational Financial Conditions index*Note: See note to part A. Sources given in the main body of the text. Generally samples begin in 1998Q1 and end anywhere from 2013.4 to 2014.2. Daily and monthly data converted to quarterly via arithmetic averaging. Generally real indicators are available at the quarterly frequency while financial and monetary policy indicators are available at higher sampling frequencies.* NFCI is primarily a financial indicator. “The Chicago Fed’s National Financial Conditions Index (NFCI) provides a comprehensive weekly update on U.S. financial conditions in money markets, debt and equity markets, and the traditional and “shadow” banking systems. Because U.S. economic and financial conditions tend to be highly correlated, we also present an alternative index, the adjusted NFCI (ANFCI). This index isolates a component of financial conditions uncorrelated with economic conditions to provide an update on financial conditions relative to current economic conditions.” From . Principal Components Analysis: ChinaFactor Loadings for Real Economic and Financial Factors (excluding monetary policy variables)Rotated loadings: L * inv(T)’F1F2Real GDP growth?0.40-0.22Cons. Price Infl?0.83-0.00Comm. Price Infl?0.89-0.03Credit growth-0.07?0.86Property Price Growth?0.04?0.45Business Climate Index?0.66?0.51Consensus 1 yr ahead growth fcast?0.24?0.70Rate of ch. Stock prices?0.45?0.32Econ. Pol. Uncert.-0.36-0.15Curr Acct/GDP ratio?0.08?0.68Energy Prod. Growth?0.67?0.35Forex Res. Growth?0.30?0.61Note: Factor loadings are based on the rotated version where the estimates are restricted to 2 estimated factors. F1 is assumed to represent the real economic factor as most of the loadings are heavily weighted on real economic variables. These are indicated in bold characters. F2 represents the financial factor since the largest loadings are dominated by financial variables. The above version excludes monetary policy variables. The relevant financial variables are in italics. All the real economic variables highlighted are positively related so that higher values indicate improving real economic conditions. The only exception is for Economic Policy Uncertainty since higher uncertainty implies poorer economic outcomes. Similarly, all the highlighted financial variables are positively related. Factor Loadings for Monetary Policy Variable (restricted to 1 factor so no rotation)LoadingsF1Money Base growth-0.19Required Reserve req.?0.52Real eff. Exch. rate?1.0024 HR PBOC rate?0.23In line with much of the relevant literature the weights are suggestive of the relative importance of the exchange rate, followed by the required reserve ratio, as the main instruments of monetary policy. The negative sign for the monetary base is because a tightening implies lower growth in contrast with higher values for the other variables.Principal Components Analysis: USAFactor Loadings for Real Economic and Financial Factors (excluding monetary policy variable)Rotated loadings: L * inv(T)'F1F2Real GDP Growth?0.63??0.26PCE deflator Inf?0.06?0.73Oil prices Inf??0.22?0.98Credit Standards index-0.78?-0.19Volume of Comm. Loans-0.05?0.26Consensus Real GDP growth fcst??0.72?0.31Term spread-0.11-0.17USA_NFCI-0.97-0.07Economic Pol. Unc.-0.37-0.24Natl Housing Price growth??0.80?0.12Note: See the Table for China for interpretation and details. Real and Financial Factors for the U.S.Additional Impulse Responses for real GDP Growth Not Included in Figure 1 for ChinaAdditional Impulse Responses for Inflation Not Included in Figure 1 for ChinaImpulse Response Functions for the Estimates in Table 1Note: RGDPCEICG is real GDP growth, INFL is CPI inflation, REERBISG is the rate of change in real effective exchange rate, EXTMSIMFAG is base money growth, COMMINFL is commodity price inflation, RRR is reserve requirement ratio, CREDITFIG is credit growth, and CHN signifies that the data are for China. Other details about the data are given above in this Appendix. The VAR includes the 2 exogenous US factors explained in the text.Version of Figure 2 Using Equation (3)Plot of Raw Series: Potential Exogenous Variables for ChinaPlot of Raw Potential Endogenous Variables for ChinaSensitivity Analysis of the Monetary Policy Factor for ChinaWe use the first difference of the monetary policy factor for China in the VAR estimation to ensure the variable is stationary. Alternatively, we could remove trends from the monetary variables prior to calculating the monetary policy factor. We thank an anonymous referee for suggesting this approach. Below we show four experiments: (1) detrend all four monetary variables via a HP filter before obtaining the monetary policy factor for China; (2) detrend just the required reserve ratio via a HP filter and then compute the monetary policy factor for China; (3) remove linear or broken linear trends from all four monetary variables before obtaining the monetary policy factor for China; (4) remove a broken linear trend from the required reserve ratio and then compute the monetary policy factor for China. We find that these monetary policy factors are highly correlated to each other and with our original monetary policy factor (which is obtained without any detrending). We also get very similar results as we estimate equation (6) using these alternative monetary policy factors. In conclusion, our overall findings are robust to the various constructions of monetary policy factor for China.The following figures show the trend and cyclical components of each of the four monetary variables after we apply the HP filters. Alternatively, we can remove a linear or broken linear trend from each of the four monetary variables. The following figures show the actual, fitted and residual values of each of the four monetary variables in this case. Notice that the break point is chosen to be 2006Q2 for the required reserve ratio and 1999Q4 for the 24 HR PBOC rate. Plot the residual values of all four monetary variables together, we haveAs shown in the above figures, different detrending methods yield similar cyclical movements (or residuals) in each monetary variable. Of course, the results based on HP filters are smoother, which is not surprising. Let m_cn represents the original monetary policy factor for China obtained using the levels of all four monetary variables. Let m_cn_hp represents the monetary policy factor for China obtained using the cyclical components (via the HP filter) of all four monetary variables. Let m_cn_hp_alt represents the monetary policy factor for China obtained using just the cyclical component (via the HP filter) of the reserve requirement ratio and the levels of the other three monetary variables. Let m_cn_lt represents the monetary policy factor for China obtained using the residuals (via a linear detrending) of all four monetary variables. Let m_cn_lt_alt represents the monetary policy factor for China obtained using just the residuals (via a linear detrending) of the reserve requirement ratio and the levels of the other three monetary variables. The following figure compares the various monetary policy factors for China.Clearly these monetary policy factors are highly correlated. The correlation coefficients are CorrelationCoefficientm_cnm_cn_hpm_cn_hp_altm_cn_ltm_cn_lt_altm_cn10.480.470.350.39m_cn_hp0.4810.750.640.68m_cn_hp_alt0.470.7510.810.96m_cn_lt0.350.640.8110.85m_cn_lt_alt0.390.680.960.851Factor Loadings for monetary policy variable (restricted to 1 factor so no rotation)Loadingsm_cnm_cn_hpm_cn_hp_altm_cn_ltm_cn_lt_altMoney Base growth-0.19-0.52-0.15-0.08-0.1Required Reserve ratio0.520.450.550.320.45Real Eff. Exch. rate1.000.370.360.420.3124 HR PBOC rate0.230.760.671.000.77Next we show that whether detrending monetary variables (via either a HP filter or broken linear trend) before obtaining the monetary policy factor for China does not affect our overall conclusions. When we use m_cn_hp in estimating equation (6), the results are essentially the same as those reported in the paper. The corresponding Table 2 and Figure 5 when using m_cn_hp are given below.Table 2 Spillovers from U.S. Real and Financial Factors to China, m_cn_hpReal Factor CHINAFinancial Factor CHINAReal USA Factor (-1)?0.16-0.05??(0.09)?(0.13)?[ 1.74385][-0.37191]Real USA Factor (-2)-0.11-?0.01??(0.09)?(0.13)?[-1.23919][ -0.06870]Financial USA Factor (-1)-0.16-0.15??(0.06)?(0.08)?[-2.60159][-1.74566]Financial USA Factor (-2)-0.01-0.12??(0.07)?(0.10)?[-0.16019][-1.21760]Figure 5 Impulse Responses Based on Equation (6), m_cn_hpWhen we use m_cn_hp_alt in estimating equation (6), the results are again similar (especially when comparing the impulse responses) but somewhat less significant than before. The corresponding Table 2 and Figure 5 when using m_cn_hp_alt are given below.Table 2 Spillovers from U.S. Real and Financial Factors to China, m_cn_hp_altReal Factor CHINAFinancial Factor CHINAReal USA Factor (-1)?0.15-0.06??(0.09)?(0.13)?[ 1.61526][-0.48968]Real USA Factor (-2)-0.110.002??(0.09)?(0.12)?[-1.23466][0.01764]Financial USA Factor (-1)-0.08-0.06??(0.07)?(0.09)?[-1.16943][-0.66790]Financial USA Factor (-2)-0.05-0.17??(0.07)?(0.10)?[-0.69723][-1.65497]Figure 5 Impulse Responses Based on Equation (6), m_cn_hp_altWhen we use m_cn_lt or m_cn_lt_alt in estimating equation (6), the results are once more similar to those reported in the paper (especially in the case of m_cn_lt) but somewhat less significant. The corresponding Table 2 and Figure 5 when using m_cn_lt or m_cn_lt_alt are given below.Table 2 Spillovers from U.S. Real and Financial Factors to China, m_cn_ltReal Factor CHINAFinancial Factor CHINAReal USA Factor (-1)?0.17-0.04??(0.09)?(0.13)?[ 1.82681][-0.29830]Real USA Factor (-2)-0.11-?0.01??(0.09)?(0.13)?[-1.22361][ -0.08802]Financial USA Factor (-1)-0.11-0.11??(0.06)?(0.09)?[-1.65237][-1.20589]Financial USA Factor (-2)0.002-0.11??(0.07)?(0.10)?[0.02304][-1.10158]Table 2 Spillovers from U.S. Real and Financial Factors to China, m_cn_lt_altReal Factor CHINAFinancial Factor CHINAReal USA Factor (-1)?0.17-0.04??(0.09)?(0.13)?[ 1.81807][-0.27395]Real USA Factor (-2)-0.13-?0.02??(0.09)?(0.13)?[-1.42780][ -0.18729]Financial USA Factor (-1)-0.10-0.10??(0.07)?(0.10)?[-1.49015][-1.02043]Financial USA Factor (-2)-0.03-0.14??(0.07)?(0.10)?[-0.37889][-1.35024]Figure 5 Impulse Responses Based on Equation (6), m_cn_ltFigure 5 Impulse Responses Based on Equation (6), m_cn_lt_alt ................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.