The simple monetarist model (inflation as a function of money supply growth and real GDP growth) seems to have been followed by the RBI in its

Size: px

Start display at page:

Download "The simple monetarist model (inflation as a function of money supply growth and real GDP growth) seems to have been followed by the RBI in its"

Ronald Thompson
5 years ago
Views:

1 Inflation, Growth and Monetary Policy Surjit S Bhalla, Jan 9, 2009 Prepared for NBER Neemrana Conference, Jan 10-13, 13, 2009

2 What determines inflation a historical view The simple monetarist model (inflation as a function of money supply growth and real GDP growth) seems to have been followed by the RBI in its monetary policy. Table 1 presents this simple model for different time-periods and specifications (either in log levels or log changes) The model works in log levels, but completely collapses in log changes; in the latter correct specification (because of strong co-integration in the levels model) money supply change is not even significant at the 10 % level of confidence Other time-period regressions suggest that the period lends support to the monetarist hypothesis; if these global inflation years are excluded, monetarism fails. Fiscal deficits equally irrelevant for determining inflation (when significant, deficits add to growth) The biggest determinant of inflation world inflation

3 Does money supply growth explain inflation? Monetarist Models of Inflation Levels (Rangarajan 1998) Log changes All log changes M3 0.35*** ** 0.41 GDP ** ** R No. of obs Notes: Other variables in the model include dummy variables for 1974 and 1980 and lagged inflation (cols 1 and 2); columns 3-5 have lagged money supply.

4 Do Fiscal Deficits Affect Growth? Dependent variable:gdp growth Fiscal Deficit lagged (% of GDP) * Real interest rate (lagged) -0.25*** -0.38*** Currency Undervaluation (lagged) -0.01* *** R Notes: 1. *Stars indicate the level of statistical significance; 1,2, and 3 stars indicate significance at the 10%, 5% and 1% level of significance respectively 2. Other variables in the equation are rainfall, rainfall lagged, and in growth equations, a dummy variable for 1991 when applicable 3. Currency undervaluation is measured as follows: RER = 1.185*( Y ) where Y is per capita per day income in 1996 PPP prices and RER is the real exchange rate conventionally defined as the ratio of PPP and US$ exchange rates. The (log) deviation from the predicted value is the level of undervaluation.

5 Domestic and World Inflation Dependent Variables: Inflation (GDP deflator) Variables World Inflation 0.32* 0.38** 0.68*** 0.35*** 0.28*** Undervaluation USA *** -0.07*** Undervaluation India *** 0.014*** R *Stars indicate the level of statistical significance; 1, 2, and 3 stars indicate significance at the 10%, 5%, and 1% level of significance, respectively. Note. Model 3, has a dummy variable for crisis year 1991.

6 Inflation: World(median) and India (GDP deflator measure) World India year

7 What determines growth a historical view Role of currency undervaluation Role of real interest rates the missing link in policy Real interest rates explains growth that did not occur, and growth that did

8 Determinants of GDP growth Coefficient of Constant Lagged Currency Undervaluation Lagged Real Interest Rate R Model ** 0.59 Model *** -0.33*** Model *** 0.43 Model *** -0.62*** 0.89 *Stars indicate the level of statistical significance; 1, 2, and 3 stars indicate significance at the 10%, 5%, and 1% level of significance, respectively Real interest rate is defined as the difference between a weighted average of government securities rate and inflation as measured by the GDP deflator.

9 The (lagged) effect of interest rates on GDP growth GDP Growth Lagged Real Interest Rates year

10 Partial regression plot of real interest rates and GDP growth, e( ygdpk X ) e( rrgsecl X ) coef = , (robust) se = , t = -7.23

11 Currency Undervaluation and GDP growth, e( ygdpk X ) e( uvl X ) coef = , (robust) se = , t = -3.06

12 Monetary Policy in the Repo period Two views should be predictable (inflation targeting) and not Repo started in 2000 Q4 under Jalan; Jalan period 2000 Q4 to 2003 Q3 (period I) Reddy period 2003 Q3 to 2008 Q3 (period II) Stark differences in monetary policy Does GDP growth matter in the setting of interest rates? No. Was the Taylor rule followed? No Did the inflation rate matter in the setting of repo rates? No Did money supply growth matter in the setting of repo rates? Yes, and only money supply growth

13 Repo Rate Determined! Repo Rate Determinants: A Simple Model (Quarterly data) (1) (2) (3) (4) Joint Jalan-Reddy period (Nov 2000 to Dec 2007) Reddy period (Sept 2003 to Dec 2007) All determinants Only money supply All determinants Only money supply Lagged % change GDP -0.22*** Lagged % change WPI Lagged % change M3 Constant *** 0.24*** 0.24*** 0.24*** 4.90*** 3.67** 3.22* 3.0*** Adjusted R -squared Observations Note: 1) Absence of * indicates not significant even at 5% level of confidence; * indicates significant at 5%, ** at 1% and *** at.1% 2) The Reddy period ended in September 2008 but the model was estimated till the end of 2007 to obtain out-of-sample predictions for 2008 as reported in the graph 3) The repo rate was introduced in November 2000

14 Actual and Predicted Repo Rate, Quarterly Data, Model 4 (based on money supply growth only)

15 Monetary Policy in the Crisis Period, 2008 Perfect fit of simple monetary model Most significant outlier Repo increase in July 2008 should have decreased!

16 Do other interest rates follow the repo rate? Yes Co-efficient of repo rate R 2 Call rate 1.31*** 0.18 Lending rate 0.58*** 0.50 Deposit rate 1.25*** 0.75 Gsec 1 year 1.07*** 0.67 Gsec 5 year 1.14*** 0.65 Gsec 10 year 1.04*** 0.72 Notes: 1. Quarterly data (start of repo) to 20084, 33 observations 2. *Stars indicate the level of statistical significance; 1, 2, and 3 stars indicate significance at the 10%, 5%, and 1% level of significance, respectively

17 Do other countries follow the India model of repo determination? Country Coefficient on lagged M3 R 2 Australia (1.9) Canada (1.08) Czech Republic (2.0) Denmark (6.23) European Monetary Union (1.42) Korea (2.13) Japan (0.97) Mexico (0.58) New Zealand (0.95) Russia (-3.2) South Africa (-3.33) UK (-2.32) USA (1.42)

18 How unusual is India Negative yield curve in the period of the Great Depression II! Very unusual pattern not followed in the US, UK, or ECB. Out of 21 countries, only 10 show a significant effect of money supply, and five have the wrong (negative) sign; for the positive coefficient countries, the significance declines when exchange rate change is added to the equation. In India, exchange rate change has had the opposite effect to that in other countries i.e. exchange rate appreciation has led to a tightening of monetary policy (higher interest rates!). IMF does not even publish money supply data in its WEO database

19 What should monetary policy be in India? How does the repo rate affect other rates? Enormously Market rates in boom and busts Rates in Developed and Emerging economies Bottom line: Real Short rates 2 to 3 percent higher in boom times, 0 to 2 in bust times Where should repo be today 3 to 4 percent

20 The mid-nineties episode The provocation inflation surge in the WPI Monetary response some parameters Other inflation indicators no follow through Impact on growth a steep decline from the new potential Growth parameters same as the no-reform 1980s

India s Economy: Performances and Challenges

India s Economy: Performances and Challenges Essays in Honour of Montek Singh Ahluwalia Indian Economic Growth: Three Puzzles Presented by Surjit S. Bhalla* February 10 th, 2010 *O[x]us Research & Investments,