K Suranjit Deputy Director, Bangladesh Bank (BB) [The views expressed here do not necessarily reflect those of BB.]
Outline Background Research Question Literature Review Data Methodology Findings Conclusion and Limitation
Background Monetary policy of Bangladesh concerns about inflation, growth, exchange rate, interest rate and money supply (BB, 2016); Contractionary policy shock increases real interest rate, appreciates domestic currency, and decreases both inflationary expectations and output (Bhuiyan, 2013); Increasing the policy interest rate decreases output, and appreciates local currency and increases the price level (Alam 2015); Monetary policy plays a better role than fiscal policy in accelerating the economic growth in Bangladesh (Maroney et al , 2004; Hasan et al., 2016) .
Background Contd. For monetary policy transmission mechanism the output decomposition technique: explains the insight of economic traits, identifies the underlying effective procedures of the monetary policy shocks to the particular real sectors (Angeloni et al. 2003; Fujiwara 2004; Phan 2014). So far as I found, still there is no research about this method for Bangladesh.
Research Question What is the more effective channel between the consumption and investment for the monetary policy transmission mechanism in Bangladesh?
Literature Review Monetary policy affects the real economy mainly through the interest rate and exchange rate (Obstfeld & Rogoff, 1995); In Bangladesh monetary policy shocks transmit to the real output through both the interest rate and exchange rate, however, interest rate instantaneously responds to exchange rate (Bhuiyan 2013); Exchange rate fluctuation affects the inflation in Bangladesh (Akhtaruzzaman, 2005).
Literature Review Contd. The investment channel is superior in the Euro area, whereas, consumption channel plays the major role in the USA (Angeloni et al., 2003) ; Investment channel works better in Japan (Fujiwara, 2004) and Australia (Phan 2014). The household spending is still the main contributor to real GDP growth in the USA (Fischer 2016);
Data Yearly data from 1973 to 2015 is used; Nine variables: the household consumption, investment, import, GDP deflator, total bank credit and central bank’s policy rate are collected from IMF (2016) while exchange rate is from WB (2016); “Other GDP” (remaining) is calculated by deducting the consumption, investment and import from GDP; India’s GDP data is also from IMF (2016).
Data Description Trend of investment data (IMF, 2016) as a percentage of the GDP is almost same as central bank’s historical data (BB 2016) available from the year 1989-90. Consumption data (IMF, 2016) is on average 5 percent lower than the central bank's data (BB 2016) , while for consistency , IMF’s data is used. Consumption data usually contains some seasonality effects; to address this issue and for easy interpretation, variables ( except the interest rate) are transformed to the logarithmic form.
Data Description Contd. The nature of the output decomposition data of Bangladesh is quite different from Australia, Japan, and the USA's data; The "other GDP" (after deducting “consumption” and “investment”) component which is used in the literature is positive ; For Bangladesh, "other GDP" becomes negative for some years with a very high standard deviation. Negative value generates missing figures when transforming in the logarithmic term, hence, another variable, the import (im) is used.
Methodology and, Two recursive (Christiano et al. 1999)VAR models (Specification 1 & 2) are used to examine the sensitivity of the investment and consumption for both the interest and the exchange rate shocks. Two non-recursive (Sims and Zha, 1998 and Kim and Roubini, 2000) identification models (Specification 3 & 4) are also used by imposing restrictions on the contemporaneous parameters considering the conditions of the economy.
Methodology Contd. The relationship between the reduced form errors and the structural disturbances for specification-1 (S-1) would be shaped as the following matrix form:
Methodology Contd. The structural matrix for the specification-4 (S-4) is constructed as follows:
Methodology Contd. Granger causality test confirms that exchange rate shock is statistically significant (rejects the null hypothesis) for the fluctuation in the consumption and price level both at 0.1 percent significant level while for the total credit the null hypothesis is rejected at 10 percent significant level; Considering the sample size and frequency one lag is used following the SBIC. (Following Lutkepohl, 1985 and Enders, 2010)
Methodology Contd. The models satisfy the stability condition as all the eigenvalues lie inside the unit circle; 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0 -0.5 -0.5 -1.0 -1.0 -1.5 -1.5 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
Findings: Response to Interest Rate Shock (S1) Consumption Investment Import .04 .04 .02 .02 .00 .00 .00 -.02 -.04 -.02 -.04 -.04 -.08 -.06 -.06 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Other GDP DGP Deflator Policy rate .15 .04 .8 .03 .10 .6 .02 .05 .4 .01 .00 .00 .2 -.01 -.05 .0 -.02 -.10 -.03 -.2 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Toal Credit .04 .02 .00 -.02 -.04 -.06 1 2 3 4 5 6 7 8 9 10
Findings: Response to Exchange Rate Shock (S2) Consumption Investment Import .04 .06 .04 .02 .04 .02 .00 .02 .00 -.02 .00 -.02 -.04 -.02 -.06 -.04 -.04 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Other GDP GDP def lator Policy Rate .12 .04 .3 .2 .08 .02 .1 .04 .00 .0 .00 -.02 -.1 -.04 -.04 -.2 -.08 -.06 -.3 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Total Credit Exchange Rate .08 .06 .06 .04 .04 .02 .02 .00 .00 -.02 -.04 -.02 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Findings: S1 with 4 lag Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LCONS to LR Response of LI to LR Response of LIM to LR .04 .08 .04 .02 .02 .04 .00 .00 .00 -.02 -.02 -.04 -.04 -.04 -.08 -.06 -.06 -.08 -.12 -.08 5 5 5 10 15 20 25 30 10 15 20 25 30 10 15 20 25 30 Response of LY to LR Response of LP to LR Response of LR to LR .02 .08 .12 .01 .08 .04 .00 .04 .00 -.01 .00 -.04 -.02 -.08 -.03 -.04 -.12 -.04 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 Response of LTC to LR .04 .02 .00 -.02 -.04 -.06 5 10 15 20 25 30
Findings: S2 with Exogenous Variable Response to Cholesky One S.D. Innovations ± 2 S.E. Response of LCONS to LEX Response of LI to LEX Response of LIM to LEX .04 .04 .03 .02 .02 .02 .01 .00 .00 .00 -.02 -.01 -.02 -.04 -.02 -.06 -.04 -.03 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 10 10 Response of LY to LEX Response of LP to LEX Response of R to LEX .12 .04 .3 .2 .02 .08 .1 .00 .04 .0 -.02 -.1 .00 -.04 -.2 -.04 -.06 -.3 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Response of LTC to LEX Response of LEX to LEX .06 .06 .05 .05 .04 .04 .03 .03 .02 .02 .01 .01 .00 .00 -.01 -.01 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 10
Finding: Variance Decomposition of S2 Consu Investm Other Policy Total GDP Exchang Variables Import Deflator e rate mption ent GDP Rate Credit 55.25 6.85 2.21 5.24 1.22 0.72 17.14 11.37 Consumption Investment 61.59 13.09 10.29 6.87 2.26 1.87 3.91 0.12 19.29 2.1 61.27 1.34 2.97 4.68 7.22 1.13 Import Other GDP 36.61 1.75 3.14 46.22 8.02 0.31 1.28 2.67 Deflator 9.37 3.71 2.36 46.52 8.17 2.53 22.17 5.18 Policy Rate 47.5 1.42 0.4 18.89 1.94 26.99 2.24 0.63 0.16 0.77 0.14 51.8 6.96 3.4 28.12 8.65 Total Credit Exchange 24.97 4.57 1.66 7.49 2.05 4.23 5.1 49.94 rate The raw variables’ percentage explained by column variables.
Findings: Responses to Exchange Rate Shock (non recursive) (S4) Consumption Investment Import .12 .12 .12 .08 .08 .08 .04 .04 .04 .00 .00 .00 -.04 -.04 -.04 -.08 -.08 -.08 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Other GDP GDP Deflator Policy Rate .4 .12 .6 .3 .4 .08 .2 .2 .04 .1 .0 .0 .00 -.2 -.1 -.04 -.4 -.2 -.3 -.08 -.6 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Total Credit Exchange Rate .3 .08 .2 .06 .1 .04 .0 .02 -.1 .00 -.2 -.02 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 10
Recommend
More recommend
