This paper discusses the intermediation of financial saving in India and the implications for growth. Recent studies linking financial sector development and growth in India are reviewed. The following statistical data are also provided: employment and labor statistics, agricultural production and yields, index of industrial production, saving and investment, price developments, balance of payments, official reserves, reserve money, monetary survey, central and state government operations, indicators of financial system soundness, financial performance of Indian commercial banks, and selected monetary and exchange rate indicators.

Abstract

This paper discusses the intermediation of financial saving in India and the implications for growth. Recent studies linking financial sector development and growth in India are reviewed. The following statistical data are also provided: employment and labor statistics, agricultural production and yields, index of industrial production, saving and investment, price developments, balance of payments, official reserves, reserve money, monetary survey, central and state government operations, indicators of financial system soundness, financial performance of Indian commercial banks, and selected monetary and exchange rate indicators.

V. Monetary Policy Transmission in India1

A. Introduction

1. The conduct of successful monetary policy critically hinges on an assessment of the timing and the effects of different shocks on important macroeconomic variables. Monetary policy is transmitted to variables such as output, prices, and the exchange rate through several channels. While these channels may not be mutually exclusive, the relative importance of each channel differs across economics depending on the underlying structural characteristics, state of development of financial markets, monetary policy instruments, the fiscal stance and the degree of openness.

2. In India, the monetary policy framework underwent a substantial transformation during the 1990s. In 1998, the Reserve Bank of India (RBI) announced a move away from a broad money target toward a “multiple indicators” approach to the conduct of monetary policy. Under the multiple indicators approach, movements in a number of macroeconomic variables, including the interest rate, exchange rate, and inflation rate are evaluated to formulate monetary policy. In general, the objective of monetary policy is to maintain price stability while ensuring adequate liquidity to meet credit growth and support investment demand in the economy.

3. This paper analyzes the channels of monetary transmission in India. Specifically, the channels examined are, (i) the interest rate channel operating through the impact of monetary policy on the cost of capital and domestic demand, (ii) the exchange rate channel which takes into account the increasing degree of openness and thus any impact on trade and capital flows, and (iii) the credit channel, in particular the bank-lending channel, focuses on the possible effects of monetary policy actions on the supply of loans by financial institutions. The interactions between real and nominal variables in the economy implied by the various channels of the transmission mechanism are analyzed empirically by using vector auto-regressions.

4. The main findings of this paper are:

  • The impact of shocks on key macroeconomic variables is larger when the exchange rate is introduced into the model. This implies that for a better understanding of monetary transmission in India, particularly since the 1990s, the exchange rate should be included in the analysis.

  • There is little evidence that bank lending plays a very important role in transmitting monetary policy changes.

  • Decomposing the macroeconomic variables into their permanent and transitory components suggests that the stance of monetary policy is broadly appropriate at the present time.

  • When the sample is split into pre- and post-reform periods, the results suggest that, (i) the wide-ranging administered prices in the 1980s seem to have limited the long-run effects of monetary expansion and tightening on prices, and (ii) the response of the nominal interest rate in the pre-reform period is smaller than that in the post-reform period, reflecting the regulated interest rate environment during the 1980s.

  • A regional comparison of the monetary policy transmission shows that monetary tightening corresponds to a pronounced increase in the real interest rate in India, Indonesia, and the Philippines, the countries that have large fiscal deficits and have experienced periods of relatively high inflation.

5. The rest of the paper is organized as follows. Section B describes the methodology adopted, the data and the sample identified in the empirical analysis. Section C discusses the results and policy implications, and Section D concludes.

B. Methodology and Data

6. The channels of monetary policy transmission are examined using a structural vector error correction model, also known as the common trends model (CTM)2. Studies that aim at evaluating the effects of monetary policy on the rest of the economy have, in general, utilized the standard reduced-form VAR framework. This framework either avoids issues of nonstationarity of the data, cointegration, simultaneity and exogeneity, or chooses recursive identification schemes that do not always have a clear economic interpretation. The CTM helps to overcome these limitations,3 by allowing the imposition of identification restrictions that are suggested by economic theory. It also deals more systematically with the issues of nonstationarity and cointegration.4 The CTM helps to test (i) hypotheses regarding long-run equilibria; (ii) the mechanisms of propagation of shocks (impulse responses); (iii) the causes of short-run fluctuations in key variables (variance decomposition); and (iv) the decomposition of endogenous variables into permanent and transitory components.

7. The CTM focuses on the interaction between real variables (output, a measure of real cash balances, and the real effective exchange rate) and nominal variables (interest rates, and inflation). The seasonally adjusted index of industrial production yt proxies economic activity (output), year-on-year change in the wholesale price index (WPI)5 πt, represents inflation, real cash balances is the M2/WPI rmt, the call money rate6 is the policy rate it, and the real effective exchange rate REERt is the exchange rate measure. Except interest rates, all data are in logs.7 Given the importance of oil prices to the Indian economy, the year-on-year change in the oil price index is included as an exogenous variable.

8. Data are quarterly from 1981:1 to 2002:4. To assess the potential changes in the transmission channels that could have occurred due to the structural reforms initiated in the economy during the 1990s, the empirical analysis is re-applied to two distinct sub-samples, 1981:1–1990:3 and 1992:1–2002:4.8

C. Empirical Evidence

9. Two sets of structural vector autoregressions are used for the analysis. The first excludes the exchange rate and thus focuses on the interest rate channel for the transmission of monetary policy. The second includes the exchange rate and thus can be used to identify the exchange rate channel. Given the key macroeconomic variables of interest—yt, rmt, REERt, it, and πt—the long-run equilibrium relationships considered in the paper are the money demand, and the Fisher parity.9 When applied to the data, estimates of the money demand function for both the interest rate and exchange rate channels are in line with previous studies on India. However, the coefficient on inflation in the Fisher relationship is lower (and below unity) for the exchange rate channel compared with the interest rate channel. Estimates below unity imply substantial adjustment in the real interest rate in response to changes in anticipated inflation.10

Long-Run Equilibrium Relationships

article image

10. The VARs are subject to a range of shocks.

  • An unanticipated monetary policy shock: A transitory shock which has no immediate effect on output; affects real and nominal variables in the short-run but not in the long run; defined as an increase in the real interest rate determined by a temporary deviation from the Fisher parity.

  • An aggregate supply shock: A permanent shock which affects both real and nominal variables in the short-run and the long run; defined as a one standard deviation in shock to output.

  • An aggregate demand shock: A transitory shock which affects real and nominal variables in the short-run; defined as a temporary deviation from the money demand relationship.

  • An inflation objective shock: A permanent shock which affects both real and nominal variables in the short-run but only the latter in the long run (long-run neutrality of money); defined as a one standard deviation shock to the monetary aggregate.11

  • The real effective exchange rate shock: A permanent shock which affects all variables in the short-run and the long run but is not affected by any of them in the long run; defined as a one standard deviation shock to the real effective exchange rate.

The Interest Rate Channel

Impulse Responses

11. An unexpected tightening in monetary policy, i.e., an increase in the real interest rate,12 leads to temporary decreases in output and inflation. Output falls reaching its trough in the second quarter. The shock is persistent with the contraction in output lasting almost 20 quarters. The response of inflation is immediate as it declines by 3½ percent in the first quarter. The nominal interest rate increases by 140 bps following the monetary tightening. The effect, however, is short-lived as the interest rate starts to fall in the fourth quarter. Initially, real cash balances increase (due to the decrease in inflation) and reach the long-run equilibrium level after 17 quarters.

uA05fig01

Response of Output to an Unanticipated Monetary Tightening Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

uA05fig02

Response of Inflation to an Unanticipated Monetary Tightening Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

12. A positive supply shock, which causes output to rise above its potential level, leads to a lower long-run inflation level In India, cyclical changes in economic activity have often been induced by supply shocks, predominantly from international oil prices. An aggregate supply shock causes an immediate and sharp decrease in inflation, over 2 percent. The shock is persistent, as inflation reaches its new long-run equilibrium (-0.2 percent) after 20 quarters.

uA05fig03

Response of Inflation to an Aggregate Supply Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

13. A positive demand shock, causes output to rise above its potential level and this translates into higher inflation (1½ percent). The monetary authorities respond by immediately raising the nominal interest rate by almost 35 bps in the first quarter and further by 125 bps in the second quarter. Recall that the dynamics between inflation and the interest rate is determined by the Fisher parity.

14. An increase in the inflation objective leads to a temporary decrease in output, and a permanent increase in the nominal interest rate. Following a permanent 1.5 percentage point increase in inflation, output falls temporarily and slowly returns to its potential level. This demonstrates the trade-off between inflation and output.13 The implied permanent increase in nominal interest rate is 140 bps.

uA05fig04

Response of Inflation to the Inflation Objective Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

uA05fig05

Response of Interest Rate to the Inflation Objective Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

Variance Decompositions

15. The variance decomposition separates the variation in an endogenous variable into the component shocks in the model and provides information about their relative importance. Results from the variance decomposition of output and inflation are discussed below.

16. Aggregate demand shocks are the main causes of output variability in the short-to medium-term (80 percent and 57 percent respectively), while in the long-term aggregate supply shocks are the main determinants (52 percent) of output variability.14 Forecast error decomposition of output is in line with model predictions, productivity (aggregate supply) developments drive trend output and aggregate demand shocks cause output to deviate from its trend. In the Indian economy, supply shocks are the main determinants of trend output.

17. In the short-term, aggregate supply shocks account for 25 percent of the inflation variability, while aggregate demand shocks account for about 10 percent of the variability. The result suggests that determining the monetary policy response to observed variations in inflation in India is not easy. In the medium- to long-term, the inflation objective shock is the main source of inflation variability, reflecting the effects of excess money growth on the economy in the long-term.

18. Decomposing the nominal interest rate into its permanent and cyclical components suggests that the stance of monetary policy is broadly appropriate at the present time. The permanent component is generated by the dynamics between the variables in the model. As shown in the figure, the permanent component of the nominal interest rate is only slightly below the actual rate.15

The Exchange Rate Channel

19. There is evidence that the exchange rate plays an important role in the transmission of monetary policy. The impact of the above identified shocks on the key macroeconomic variables is magnified somewhat when the exchange rate channel is included as an additional transmission channel.16 The interest rate channel is augmented to incorporate an exchange rate measure, in this case the real effective exchange rate.

20. An unexpected tightening in monetary policy causes the real effective exchange rate to appreciate by 0.5 percent.17 The appreciation in the real effective exchange explains the larger decrease in output (0.4 percent) in the exchange rate channel of monetary policy transmission. The shock is quite persistent with the contraction in output lasting almost 30 quarters.

21. A positive aggregate supply shock causes the real exchange rate to depreciate by 1 percent. The depreciation is persistent, lasting for five quarters, and this could help explain the larger impact of this shock on variables in this model. The variance decomposition analysis shows that the contribution of the aggregate supply shock to output variability increases in the short- to medium-term (40 percent and 60 percent respectively), compared to the interest rate channel.

uA05fig07

Response of Output to an Unanticipated Monetary Tightening Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

uA05fig08

Response of REER to an Unanticipated Monetary Tightening

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

The Credit Channel

22. There is little evidence that monetary shocks are transmitted through bank-lending. The interest rate channel is augmented to include nonfood credit extended by commercial banks. This permits an examination of how changes in the policy rate affect bank-lending, and consequently, how bank-lending affects output.

uA05fig09

Response of Bank-Lending to an Unanticipated Monetary Tightening

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

Impulse Responses

23. There is weak transmission from monetary policy changes to bank-lending. Bank-lending falls in response to an unanticipated monetary tightening shock. However, the magnitude of the response is small and slow. As shown in the figure, bank lending reaches its trough after ten quarters. This can be explained by the low-level of pass-through of the changes in the policy rate to lending rates and to credit deployment. This has reduced the efficacy of the credit channel of monetary policy.18 In large part, this is due to the fact that a substantial part of the estimation period was dominated by credit budgeting through direct lending programs and heavy reserve requirements.19

Variance Decomposition

24. Output does not react to bank-lending shocks.20 Shocks to bank lending explain only 1 percent of output variability over a 12-quarter period. However, shocks to output explain 40 percent of variability in bank-lending over the same period. This suggests that demand for credit factors may be more significant in explaining credit flows than factors that tend to influence the supply of credit. Alternatively, the finding that output does not react to bank-lending shocks could be interpreted as suggestive that unsystematic lending shocks are small.21

The Pre- and Post-Reform Periods

25. After the balance of payments crisis in the early 1990s, India entered a process of wide-ranging structural reforms. These reforms aimed at liberalizing the economy, inducing competition and instilling macroeconomic discipline. A number of reforms had a direct impact of the conduct of monetary policy: deregulating the interest rate; reducing the pre-emptions of resources from the banks through cuts in the cash reserve ratio (CRR) and statutory liquidity requirement (SLR); increasing reliance on indirect methods of monetary control; moving towards universal banking; and relaxing capital controls and allowing greater exchange rate flexibility. To assess the changes in the transmission that could have occurred due to the structural reforms, the sample was split into two, pre-reform (1981:1–1990:3), and post-reform (1992:1–2002:4).22

26. The fairly wide-ranging administered prices in the 1980s seem to have limited the long-run effects of monetary expansion and tightening on prices. In response to an unanticipated monetary tightening shock, inflation falls by more in the pre-reform period compared to the post-reform period. This holds true for the two channels of monetary policy transmission, the interest rate and exchange rate channels. Similarly, in response to an aggregate supply shock, in the post-reform period, inflation is permanently higher compared to the pre-reform period. These results may reflect the greater openness in the nineties, and suggest that inflation is affected by external factors as well.

uA05fig10

The Interest Rate Channel: Response of Inflation to an Unanticipated Monetary Tightening Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

uA05fig11

The Interest Rate Channel: Response of Inflation to an Aggregate Supply Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

27. The response of the nominal interest rate in the pre-reform period is smaller than that in the post-reform period, reflecting the regulated interest rate environment during the 1980s. In response to an unanticipated monetary tightening, the nominal interest rate increases by more in the post-reform period compared to the pre-reform period. This holds true for the two channels of monetary policy transmission, the interest rate and exchange rate channels. Similarly, in response to an aggregate demand shock, in the post-reform period, the hike in the interest rate is higher compared to the pre-reform period. This implies that in the post-reform period, curbing inflationary pressures has been key in the conduct of monetary policy.

uA05fig12

The Interest Rate Channel: Response of Interest Rate to an Aggregate Demand Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

uA05fig13

The Interest Rate Channel: The Response of Interest Rate to a Monetary Tightening Shock

Citation: IMF Staff Country Reports 2003, 261; 10.5089/9781451818574.002.A005

A Regional Comparison

28. A regional comparison23 of the impact of an unanticipated monetary tightening shock, reveals a greater degree of monetary tightening in India, Indonesia and the Philippines, the countries that have large fiscal deficits and periods of double-digit inflation.

  • The interest rate channel: There is a significant increase in the real interest rates in India and Indonesia, around 500 bps, following the tightening of monetary policy. The effects of discretionary monetary policy on output and inflation are found to be short-lived in Indonesia, Singapore, and Taiwan, but rather persistent in India, Japan, and Korea. The full impact of this shock on output and inflation is reached on average after 4 quarters.

  • The exchange rate channel: The real effective exchange rate appreciates after a monetary tightening. The response of output is more persistent compared to the interest rate channel, and reaches its trough after 5 quarters. There is a significant increase in the real interest rate in the Philippines and India, around 400 bps.

D. Conclusion

29. The interest rate and exchange rate channels of the monetary policy transmission mechanism are the most important. However, the impact of shocks on key macroeconomic variables is larger when the exchange rate is accounted for. This implies that to better understand the monetary transmission in India, the exchange rate should be included in the analysis.

30. The low-level of pass-through of the changes in the policy rate on to the lending rate coupled with directed lending requirements has reduced the efficacy of the credit channel of monetary policy. The findings in this paper are broadly in line with those of RBI (2003), where it is noted that “an important factor determining the effectiveness of the monetary transmission process is the degree of pass-through. In view of the weak sensitivity of the bank lending rates to changes in the bank rate, the efficacy of the monetary policy in reinvigorating growth runs up against a constraint.” For better monetary policy transmission, more flexibility should be introduced in the commercial bank interest rate structure. A higher pass-through would lead to declines in transmission lags. Recently commercial banks were encouraged to introduce a flexible interest rate option for all new deposits and urged to review and announce the maximum spreads around the prime-lending rate.

References

  • Al-Mashat, Rania, and Papa N’Diaye,Assessing the Effects of Monetary Policy in Asia: Evidence from a Common Trends Approach,IMF Working Paper, forthcoming.

    • Search Google Scholar
    • Export Citation
  • Bernanke, Ben and Mark Gertler, 1995, “Inside the Black Box: The Credit Channel of Monetary Policy Transmission,The Journal of Economic Perspectives, Vol. 9 (4), pp. 2748.

    • Search Google Scholar
    • Export Citation
  • Blanchard, Oliver and Danny Quah, 1988, “The Dynamic Effects of Aggregate Demand and Supply Disturbances,American Economic Review, Vol. 79, pp. 655673.

    • Search Google Scholar
    • Export Citation
  • Cassola, Nuno, and Claudio Morana, 2002, “Monetary Policy and the Stock Market in the Euro Area,ECB Working Paper Series No. 119 (January).

    • Search Google Scholar
    • Export Citation
  • Crowder, William, and Dennis Hoffman, 1996, “The Long-Run Relationship Between Nominal Interest Rates and Inflation: The Fisher Equation Revisited,Journal of Money, Credit, and Banking, Vol. 28., Issue 1, (February), pp. 10218.

    • Search Google Scholar
    • Export Citation
  • Fung, Ben, 2002, “A VAR Analysis of the Effects of Monetary Policy in East Asia,BIS Working Papers, No. 119, (September).

  • Jacobsson, Tor, Per Jansson, Andres Vredin, and Andres Warne, 2002, “Identifying the Effects of Monetary Policy Shocks in an Open Economy,Sveriges Riksbank Working Paper Series, No. 134, (May).

    • Search Google Scholar
    • Export Citation
  • Johansen, S., 1991, “Estimation and Hypothesis Testing of Cointegrating Vectors in Guassian Vector Autoregressive Models,Econometrica, Vol. 59, pp. 15511580.

    • Search Google Scholar
    • Export Citation
  • Johansen, S.,Statistical Analysis of Cointegrating Vectors,Journal of Economic Dynamics and Control, Vol. 12, pp. 23154.

  • Kapur, Muneesh, and Michael Patra, 2000, “The Price of Low Inflation,Reserve Bank of India Occasional Papers, Vol. 21, No. 2 and 3, (Monsoon and Winter).

    • Search Google Scholar
    • Export Citation
  • Leeper, Sims and Zha, T., 1996, “What Does Monetary Policy Do?,Brookings Papers on Economic Activity, Vol. 2, pp. 163.

  • Mishkin, Frederic, 1995, “Symposium on the Monetary Transmission Mechanism,Journal of Economic Perspectives, Vol. 9, No. 4, (Fall), pp. 310.

    • Search Google Scholar
    • Export Citation
  • Mukhopadhyay, Hiranya, 1999, “Monetary Transmission Mechanism: The Credit Channel Hypothesis Revisited,International Journal of Development Banking, Vol. 17, No. 2, (July), pp. 4758.

    • Search Google Scholar
    • Export Citation
  • Peersman, Gert and Frank Smets, 2001, “The Monetary Transmission Mechanism in the Euro Area: More Evidence from VAR Analysis,ECB Working Paper Series, No. 91, (December).

    • Search Google Scholar
    • Export Citation
  • Rangarajan, C., and R.R. Arif, 1990, “Money, Output, and Prices: A Macroeconometric Model,Economic and Political Weekly, Vol. 25, (April) pp. 83752.

    • Search Google Scholar
    • Export Citation
  • Ray, Partha, Himanshu Joshi, and Mridul Saggar, 1998, “New Monetary Transmission Channels: Role of Interest Rates and Exchange Rate in Conduct of Indian Monetary Policy,Economic and Political Weekly, Vol. 33, (October), pp. 278794.

    • Search Google Scholar
    • Export Citation
  • Reserve Bank of India, 2003, Report on Currency and Finance, (Mumbai).

  • Shome, Parthasarathi, and Hiranya Mukhopadhyay, 1998, “Economic Liberalization of the 1990s: Stabilization and Structural Aspects and Sustainability of Results,Economic and Political Weekly, Vol. 33, (July), pp. 192534.

    • Search Google Scholar
    • Export Citation
  • Sims, C.,Macroeconomics and Reality,Econometrica, Vol. 48, pp. 148.

  • Smets, F., 1997, “Financial Asset Prices and Monetary Policy: Theory and Evidence,CEPR Working Paper No. 1751.

  • Taylor, John, 1997, “A Core of Practical Macroeconomics,American Economic Review, Vol. 87, No. 2, (May), pp. 23335.

  • Warne, Andres, 1993, “A Common Trends Model: Identification, Estimation, and Inference,Institute for International Economic Studies, Seminar Paper No. 555.

    • Search Google Scholar
    • Export Citation

STATISTICAL APPENDIX

Table 1.

India: GDP at Factor Cost by Sector of Origin, 1996/97–2002/03 1/

article image
Source: Central Statistical Organization (CSO).

Data on a fiscal year basis beginning April 1.

Includes mining and quarrying; manufacturing; electricity, gas, and water supply; and construction.

Includes trade, hotels, and restaurants; transport, storage, and communication; financing, insurance, real estate, and business services; and community, social, and personal services.

Table 2.

India: GDP at Market Prices by Expenditure Components, 1996/97–2001/02 1/

article image
Source: Central Statistical Organization (CSO).

Data are provisional.

Residuals.

Table 3.

India: Employment and Labor Statistics, 1996/97–2000/01

(In millions of persons, end-of-period)

article image
Source: CEIC.
Table 4.

India: Agricultural Production and Yields, 1996/97–2002/03

article image
Sources: Government of India, Economic Survey; and data provided by the Indian authorities.

Third advance estimate.

Nine major oilseeds.

In million of bales of 170 kg each.

In million of bales of 180 kg each.

Data are on a calendar year basis. For example, data under the heading 1995/96 are for 1995.

In tons per hectare.

Table 5.

India: Index of Industrial Production, 1996/97–2002/03

article image
Source: CEIC.

Weights are for 1993/94 base-year data. Weights for index classified by use were revised slightly for data from 1998/99 onwards.

Table 6.

India: Saving and Investment, 1996/97–2001/02 1/

article image
Source: Central Statistical Organization (CSO).

Data are provisional.

Not adjusted for statistical discrepancy.

Percent change.

Table 7.

India: Price Developments, 1996/97–2002/03

article image
Source: CEIC.
Table 8.

India: Balance of Payments 1997/98–2002/03 1/

(In billions of U.S. dollars, unless otherwise indicated)

article image
Sources: CE1C; and Fund staff estimates.

Indian authorities’ presentation except for “Other capital” or as noted.

Net foreign direct investment in India less net foreign investment abroad.

Residual-maturity basis, except medium and long-term NRI deposits, which are on a a contracted-maturity basis.