Project on Assessment of Key Issues Related to Monetary Policy
[Source: RBI Report on Currency & Finance 2003-04]

Module: 4 Monetary Policy and Inflation

Modelling Inflation in India (Contd)

With gradual opening up of the Indian economy, international prices have started to play a key role in domestic inflation. Import prices in local currency terms include not only the effect of international prices but also exchange rate movements. There are periods when the exchange rate has appreciated while import prices have increased and vice versa . Movements in import prices in rupee terms could, therefore, be on account of movements in the import prices in foreign currency or in exchange rate or a combination of both. Segregating the influences of import prices in foreign currency on domestic inflation from that of the exchange rate movements could provide additional analytical insights in understanding the inflation process. Accordingly, the equation includes import price inflation (in US dollar terms) and exchange rate separately.

In brief, the explanatory variables include: lagged inflation, output gap, foodgrains inflation, import price inflation in US dollar terms and variation in exchange rate of the Indian rupee (vis-a-vis the US dollar). The equations are estimated over the period 1970-2004 using annual data5. The explanatory power of the equation is satisfactory, ranging from 60-92 per cent6. The various diagnostic tests such as serial correlation and ARCH errors suggest that estimates are robust.

Key results emerging from these estimates are as follows. First, lagged inflation is significant although the size of the coefficient is not very large. This suggests that inflation may be persistent, i.e., a shock that raises inflation in one year will impart an upward push to inflation expectations in the year ahead and vice versa. Amongst the three measures of inflation, the persistence appears to be maximum for GDP deflator measure and the lowest for wholesale price inflation. Second, excess domestic demand conditions have the expected positive effect on inflation. An increase of one percentage point in output gap (i.e., if actual output exceeds its trend level by one per cent) raises inflation rate by 31-56 basis points with a lag of one year, depending upon the inflation measure. The effect is maximum in case of GDP deflator and the least in the case of consumer price inflation.

Third, the supply shocks emanating from foodgrains prices play an important role in the inflation process in India. Fourth, import price inflation has the expected positive effect on domestic inflation. Estimates suggest that an increase of 10 per cent in import price inflation raises domestic inflation by up to 1.1 percentage points. The effect is the minimum for CPI inflation (0.5 percentage points) followed by GDP deflator (0.8 percentage points) and wholesale inflation (1.1 percentage points). These estimates appear to be consistent with the openness of the Indian economy over the sample period - an average of around nine per cent, increasing from four per cent to 13 per cent. Finally, exchange rate depreciation has also the expected effect of raising domestic prices and the coefficient of exchange rate pass-through to domestic inflation ranges between 8-17 basis points, i.e., a 10 per cent depreciation of the Indian rupee (vis-a-vis the US dollar) would, other things remaining unchanged, increase consumer inflation by less than one percentage point and the GDP deflator by 1.7 percentage points. The empirical results throw some differences between the pass-through from import price inflation and exchange rate movements. While import prices impact on domestic inflation in the same year, the exchange rate movements seem to affect inflation with a lag of one year (two years in case of consumer inflation). Another difference is that the pass-through from exchange rates to inflation is somewhat larger than that of pass-through from import prices. This result is consistent with available empirical evidence for the EMEs. Mihaljek and Klau (2001) who undertake separate estimates of the two pass-through effects for a sample of 13 EMEs find that the exchange rate pass-through coefficient to consumer price inflation is larger than that of import price inflation in seven countries; the coefficient is the same in four countries and lower in the remaining two countries. In contrast, for a sample of nine industrial economies, McCarthy (2000) finds that the effect of import prices on inflation is stronger than that of exchange rate.

Compared to many other EMEs, pass-through coefficients in India are fairly small. This could be attributed to two key factors: the relatively low degree of openness of the Indian economy as well as low inflation rates. Most studies document that pass-through is high in a high-inflation environment.

The Indian economy has witnessed significant structural changes since the reforms process began in early 1990s. These could have impacted upon the coefficient estimates. Formal stability tests such as Andrews and Andrews-Ploberger tests suggest that parameter estimates have been stable over the sample. Rolling regressions provide an alternative approach of assessing if parameters are stable over the period. For this purpose, rolling regressions (based on a moving sample of 20 years) are undertaken. The results from this analysis can be summarised as follows. First, there is some increase in the coefficient on lagged inflation, i.e., inflation persistence appears to have increased. Second, the coefficient on output gap remains almost unchanged in the equations for consumer price inflation and GDP deflator. For WPI equation, the coefficient shows some decline, but is imprecisely estimated. The decline in output gap terms in WPI could be reflecting the increased role of movements in international commodity prices. These prices are determined in world markets and reflect global demand-supply gaps. Taken together, excess demand conditions are an important determinant of inflation and macroeconomic policies have an important role to play.

Third, the coefficient on the exchange rate exhibits a declining trend although the estimates turn out to be somewhat imprecise. This suggests a possible decline in exchange rate pass-through to domestic inflation - these results appear to be consistent with the empirical exercise undertaken in Module: 6 which shows a lower pass-through in the recent decade. The decline in pass-through during the 1990s is consistent with the cross-country empirical evidence discussed earlier. In India, as noted earlier, inflation rates have declined significantly since the second half of the 1990s and this could be one explanation for the lower pass-through . Another key factor that could have lowered the pass-through is the phased decline in tariffs as well as non-tariff barriers such as quotas. Average import duties are now less than one-third of what they were a decade ago. This steep reduction in tariffs could have easily allowed domestic producers to absorb some part of the exchange rate depreciation without any effect on their profitability.

The regression results are subject to the caveat that these are average estimates over the sample period and are based on the assumption that other factors are unchanged. The modelling approach followed in this Module would need to be validated with other approaches. Alternative indicators to capture demand pressures in the economy need to be explored. As Brave and Fisher (2004) observe, particular forecasting models which do well in some periods may perform poorly at other times and, therefore, a useful approach is to be "eclectic with respect to both the data used to formulate a forecast and the models used to incorporate the data into a forecast. Relying on a small number of inflation indicators and one forecasting model is not a good idea". A multiple indicator approach to monetary policy formulation being followed by the Reserve Bank of India since 1998 is thus an appropriate approach.

In brief, while supply shocks - both oil and food - have an important role in year-to-year inflation, expansionary fiscal policy during the 1980s and its monetisation was indeed a major cause of the increase in inflation. Structural reforms, improved monetary-fiscal interface, reforms in the Government securities markets and effective sterilisation of capital inflows enabled better monetary management from the second half of the 1990s onwards. Large food stocks and foreign exchange reserves have played an important role in supply management efforts by keeping actual inflation low and inflation expectations under control. This was amply exhibited during 2002-03 when the country faced its worst drought in nearly two decades but adequate food stocks and large reserves kept inflation as well as expectations low. This has led to a sustained reduction in inflation, notwithstanding increases in some years due to large supply shocks. Empirical exercise undertaken in this Section confirms that both demand and supply shocks have an impact on inflation. Excess demand conditions increase inflation. There is some preliminary evidence that exchange rate pass-through to domestic inflation has declined since mid-1990s.

The success with achieving and maintaining low inflation since mid-1990s has led to a number of positive developments. First, there is virtually a national consensus that high inflation is not good and that it should be brought down. Second, inflation expectations have come down and, consequently, inflation tolerance has also come down. "The society, the economic agents and the market participants are now reacting to a lower headline inflation, which is a good thing because if we want a sustained strong growth, it is necessary to rein in inflationary expectations". In this context, despite a significant improvement in the monetary-fiscal interface during the 1990s, fiscal dominance continues to persist with growing volume of gross market borrowings (RBI, 2003). Inflation expectations, inter alia, depend upon fiscal prudence. It is, therefore, essential to pursue fiscal consolidation, promptly and with resolve, from a medium-term perspective. In this context, the Fiscal Responsibility and Budget Management (FRBM) Act enacted by the Central Government is expected to provide the Reserve Bank necessary flexibility to maintain low and stable inflation. Adherence to these fiscal rules thus become important to stabilise inflation expectations.