Transmission in India:

Transcription

1 Asymmetry in Monetary Policy Transmission in India: Aggregate and Sectoral Analysis Brajamohan Misra Officer in Charge Department of Economic and Policy Research Reserve Bank of India VI Meeting of Open Market Operators Jointly organized by Centre for Latin American Monetary Studies and Banco Central de Bolivia La Paz, April

2 Outline I. Objectives and Motivation II. Inflation-Growth Dynamics in Indian Economy in the Recent Period III. Monetary Policy Transmission: Empirical Evidence in India. IV. Data and Methodology V. Asymmetric Response of Output and Inflation to Monetary Policy: Aggregate Level VI. Analysis at Sectoral Level IV. Conclusion 2

3 Motivation Increased role of central banks in macroeconomic stability Private sector: Pressure to deleverage Public Sector: Fiscal strains and sovereign concerns. EMEs have shifted monetary policy focus to interest rate based actions. Sources of price and output shocks are external and assymetric. Domestic demand management using monetary policy has become more challenging 3

4 Two important Questions Impact of policy interest rate on output and inflation, How asymmetric? Nature of asymmetry decide relative emphasis on growth and inflation. Should a central bank be more aggressive in fighting inflation or growth? Does output and inflation i across sectors respond asymmetrically to policy changes? Are sectoral output responses to monetary policy shock asymmetric? Do sources of inflation (supply/demand) determine its responsiveness to policy 4

5 Inflation-Growth Dynamics in Indian Economy in the Recent Period Robust growth in the five years preceding the crisis of 2008 averaging at about 9 per cent. Moderate inflation during most part of this period, acted as a favourable condition for high growth. Global commodity price spike of : Impact on inflation in India was substantial Wholesale price index (WPI) inflation reaching a peak of 12.8 per cent in August Monetary policy rates were increased significantly: Repo rate was raised by 300 basis points to 9 per cent, the cash reserve ratio (CRR) for scheduled banks was raised by 450 basis points to 9 per cent. 5

6 Inflation-Growth Dynamics in Indian Economy in the Recent Period The onset of Global Financial Crisis The Indian economy witnessed moderation in growth in to 6.7 per cent while, inflation declined sharply to 0.8 per cent by end-march Significant Monetary Policy easing Between October 2008 and April 2009 CRR was reduced by a cumulative 400 basis points to 5.0 per cent of NDTL, Repo rate was cut by 425 basis points to 4.75 per cent, and Access to potential liquidity provided through conventional and non- conventional windows, which aggregated to high as 10 per cent of GDP. 6

7 Inflation-Growth Dynamics in Indian Economy in the Recent Period Growth rebounded strongly in and (to 8.4 per cent in each year). Inflation spiked and remained in the range of 8 to 11 per cent over 23 successive months between January 2010 and November Monetary policy was tightened by cumulative increase in the repo rate by 375 basis points to 8.5 per cent, besides increase in CRR by 100 basis points to 6 per cent. Since December 2011, inflation has started to moderate while growth for at 6.9 per cent also shows weakness in momentum. 7

8 Monetary Policy Transmission in India: Empirical Evidence Interest rate based policy became prominent since Transmission Continues to be black-box (Bernanke and Gertler, 1995). The Working Group on Money Supply (Chairman: YV Y.V. Rdd Reddy, 1998), Singh and Kalirajan (2007), Patra and Kapur (2010) and Pandit and Vashisht (2011) found that interest rate changes do affect both output and ifl inflation, i but the extentand lagsvary significantly ifi across the studies. Mohanty (2012) analysed the impact of interest rate based policy actions on output and inflation using an SVAR framework: policy rate increases have a negative effect on output growth with a lag of two quarters and a moderating impact on inflation with a lag of three quarters. The overall impact persists through 8-10 quarters. 8

9 Data and Period of Analysis Post April 2000 quarterly data covering upto December 2011 are used as interest rate based policy action became prominent only during the last decade. d Call money rate is used as a proxy for Policy rate. GDP growth both aggregate and sectoral (agriculture, (g industry and services). Inflation (WPI) both overall and sector-wise, Food, fuel and Manufactured non-food products. 9

10 Methodology: Vector Autoregression (VAR) VAR provides an atheoretical approach. Impulse Reponses enable us to better gauge the nature of asymmetry with a time dimension. A VAR model describes the evolution of a set of k endogenous variables over the same sample period (t = 1,..., T) as a linear function of only their past evolution. The variables are collected in a k 1 vector y t, which has as the i th element y i,t the time t observation of variable y i. A (reduced) p-th order VAR, denoted VAR(p), is where c is a k 1 vector of constants (intercept), A i is a k k matrix (for every i = 1,..., p) and e t is a k 1 vector of error terms which are white noise. 10

11 Analysis at Aggregate Level For the aggregate level, the variables under study are policy rate (weighted average call money rate), output (growth rate in GDP) and prices (WPI inflation). Unit root tests suggests that all the variables are stationary at the levels. the optimal lag length for VAR estimation was chosen to be 1 (SIC). The characteristic roots were found to be within the unit root circle. LM test suggests presence of no auto correlation. 11

12 Analysis at Aggregate Level Impulse Responses Response to Cholesky One S.D. Innovations ± 2 S.E. 1.5 Response of CALL to CALL Response of CALL to TOTALGDP Response of CALL to WPI_AC Response of TOTALGDP to CALL Response of TOTALGDP to TOTALGDP Response of TOTALGDP to WPI_AC Response of WPI_AC to CALL Response of WPI_AC to TOTALGDP Response of WPI_AC to WPI_AC

13 Analysis at Aggregate Level Annex Table 2: Variance Decomposition: Overall Model Cholesky Ordering: CALL GDP WPI Variance Decomposition of CALL: Period S.E. Call GDP WPI Variance Decomposition of TOTALGDP: Period S.E. Call GDP WPI Variance Decomposition of WPI_AC: Period SE S.E. Call GDP WPI

14 Major Results: Overall Model The call money rate responds positively to both growth and ifl inflation. i Response is higher hih for inflation ifl i than growth. Policy shock has an immediate negative impact on output and the effect wanes off in about two years. Variance decomposition exercise indicates that about 23 per cent of variance in output over a period of 6 quarters could be explained by changes in policy rate. Increase in policy rate in the immediate period tend to give an upward pressure on inflation (price puzzle) but has a negative impact on inflation sets in over a period of 6 quarters (about 7 per cent). 14

15 Analysis at Sectoral Level For the sectoral level, the variables under study are policy rate (weighted average call money rate), GDP and Inflation. GDP growth both aggregate and sectoral (agriculture, industry and services). Inflation (WPI) both overall and sector-wise, Food, fuel and Manufactured non-food products. 15

16 Analysis at Sectoral Level Chart 2: Impulse Response: Sectoral Analysis Response to Cholesky One S.D. Innovations ± 2 S.E. Response of AGRI to CALL Response of INDUSTRY to CALL Response of SERVICES to CALL Response of MNFP to CALL Response of FOOD to CALL Response of FUEL to CALL

17 Major Results: Sectoral Model Output of agriculture sector is largely insensitive to monetary policy, so is the case for food inflation. The response of industrial sector output and manufactured non-food products inflation to monetary policy is quite significant. Services sector output also respond to policy actions but with a lag. About 14 per cent of variation iti in output t of each of industrial sector and services sector is explained by changes in call rate. For manufactured non-food products inflation, the impact of policy rate over a period of 8 quarters comes out to be about 14.5 per cent whereas the impact of fuel inflation on manufactured products inflation is more stronger at above 20 per cent. 17

18 Conclusions First, the impact of interest rate changes is more immediate on output than on prices in India. Second, there is evidence of inflation increasing in the immediate period before declining following an increase in interest rate. This stands as an evidence of price puzzle in India. Third, policy respond much more to inflation shock than growth shock. Fourth, sectoral level analysis indicates that the farm output is largely insensitive to changes in interest rates whereas industrial sector and services sector output respond to interest rate actions. Similarly, food inflation is insensitive to monetary policy shock, whereas core inflation responds to monetary policy. 18

19 Conclusions The study indicates the need for calibrating monetary policy on the basis of sources of price pressures. Even thoughh monetary policy actions have limitedit role in containing supply side price pressures, one has to recognize that the spillover of pressures from supply side inflation to core inflation. Monetary policy has take into account the growth- inflation dynamics and the transmission asymmetry. 19

20 Thanks 20